Prospective study of diarrheal diseases in Venezuelan children to evaluate the efficacy of rhesus rotavirus vaccine

Journal of Medical Virology 30219-229 (1990)

Prospective Study of Diarrheal Diseases in Venezuelan Children to Evaluate the Efficacy of Rhesus Rotavirus Vaccine

Irene Perez-Schael, Dorys Garcia, Marino Gonzalez, Rosabel Gonzalez, Naimeh Daoud, Mireya Perez, Walter Cunto, Albert Z. Kapikian, and Jorge Flores Instituto de Biomedicina, Universidad Central de Venezuela and Ministerio de Sanidad y Asistencia Social (I.P. -S., D.G., M.G., R.G., N.D.), Hospital de Ninos “J. M . De Los Rios” (M.P.), and Centro Cientifico, IBM de Venezuela (W.C.), Caracas, Venezuela; Laboratory of Infectious Diseases, National Institute of Allergy and Infectrous Diseases, National Institutes of Health, Bethesda, Maryland (A.Z.K., J.F.)

The efficacy of a rhesus rotavirus vaccine (MMU 18006, serotype 3) against infantile diarrhea was evaluated by active home surveillance of a group of 320 children 1-10 months of age in Ca- racas, Venezuela. During a 1 year period following oral administration of vaccine or placebo under a double-masked code, over 600 diarrheal episodes were detected. Etiologic studies re- vealed that heat-stable toxin (ST) producing en- terotoxigenic f. coli (ETEC) was the most common diarrheal agent detected (34%) followed by enteropathogenic E. coli (EPEC, 10.9%), heat-labile toxin (LT) producing ETEC (7.6%). rotavirus (6.9%), Cryprooporidiurn (4.8%) and Carnpylo- bacrer (1.3%). ST-producing ETEC were also re- covered from over 20% of control stool specimens obtained during diarrhea-free periods, whereas EPEC, rotavirus, Cryprosporidium, and Carnpylobacrer were rarely detected in such control specimens. Rotavirus was responsible for about one-half of the more severe cases of diarrhea. Twenty-two of 151 infants who received placebo (14.6%) and eight of 151 receiving a lo4 PFU dose of vaccine (5.3%) had rotavirus diarrhea during the follow-up period for an efficacy level of 64% against any rotavirus diarrhea. However, vaccine efficacy reached 90% against the more severe cases of rotavirus diarrhea and was notably high in the 1-4 month age group. Serotypic analysis of the rotaviruses detected suggests that the resistance induced by the vaccine was type specific since significant protection was only evident against serotype 3 rotaviruses. A lo3 PFU dose tested initially in 18 children did not appear to protect against rotavirus diarrhea.

KEY WORDS: gastroenteritis, immunoprophy- laxis, infants

0 1990 WILEY-LISS, INC.

INTRODUCTION Longitudinal studies based on home surveillance are

needed to formulate vaccination strategies against diarrheal diseases, which are a major cause of infantile morbidity and mortality. Because of the recognized im- portance of rotaviruses as causative agents of this syndrome, the development of a rotavirus vaccine has a high priority [Kapikian et al., 1988a; De Zoysa and Feachem, 19851. Such a vaccine could prevent the death of several hundred thousand infants and young children annually throughout the world and in addition, by decreasing the number of diarrheal episodes, might have an impact on infant nutrition. Current strategies for rotavirus vaccine development are based on the “Jennerian” principle of using antigenically re- lated animal viral strains naturally attenuated for the human host to induce cross-reactive immunity in humans [Kapikian et al., 1986, 198813; Flores and Kapikian, 1988al. The rhesus rotavirus (RRV) vaccine tested in this study is antigenically similar to human type 3 strains, a rotavirus serotype commonly found in children with diarrhea [Taniguchi et al., 1987; Flores et al., 1988133. Previous studies with a lo4 PFU dose of this vaccine in the United States [Losonski et al., 1986; Anderson et al., 1986; Christy et al., 19861, Finland [Vesikari et al., 19861, Sweden [Gothefors et al., 19891, and Venezuela [Perez-Schael et al., 19871 have shown this vaccine to be antigenic and safe (although 20-30% of the infants receiving it in Finland, Sweden, and the USA developed short-lived, mild, febrile reactions). In a preliminary communication describing the efficacy of this vaccine in 247 children included in this report the vaccine was protective, especially when administered to 1-5 month-old infants [Flores et al., 19871.

Accepted for publication December 14, 1989. Address reprint requests to Dr. Jorge Flores, Laboratory of In-

fectious Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892.

220 Perez-Schael et al.

In this report further analysis of the vaccine efficacy trial in Venezuela is presented with a detailed descrip- tion of its efficacy in 320 children, and for the first time the overall immunogenicity of the vaccine, the serotype specificity of the rotaviruses isolated, and the role of bacteria and parasites detected during the trial are described.

SUBJECTS AND METHODS Population Studied

The study was carried out in Cotiza and Los Erasos, two neighborhoods of low socio-economic status in the North of Caracas, Venezuela. Two health units administered by the Venezuelan Ministry of Health and the Hospital J.M. De Los Rios were used for vaccination and continuous surveillance activities. Volunteer mothers were asked to read and sign a consent form previously approved by the Venezuelan Ministry of Health, the Venezuelan Medical Federa- tion, and the Clinical Research Subpanel at the National Institute of Allergy and Infectious Diseases (National Institutes of Health, Bethesda, Maryland, USA), which had reviewed the study protocol. Only well-nourished infants with no debilitating conditions were enrolled.

Vaccine Administration Administration of vaccine or placebo was carried out

sequentially in four periods (February-March 1985, J u n e J u l y 1985, October 1985, and February 1986) to confirm vaccine safety in infants within a given age group (starting with 4-10 month-old infants) and pro- ceeding to progressively younger age groups (3, 2, and 1 months old).

A randomized double-masked code was employed throughout the four phases. After vaccine or placebo was given to each group of infants, an independent observer examined the data on post-“inoculation~’ reactions to decide whether the study should continue. This procedure was adopted since the vaccine had never been given previously to infants under 4 months of age. The code was partially disclosed to the participants at the National Institutes of Health before the end of the trial to analyze vaccine reactions and serological results in order to plan further studies with the RRV vaccine in this and other locations. The code was bro- ken after the tabulation of laboratory results and scoring of illness severity.

Initially, two different doses of vaccine (lo4 and lo3 PFU) were tested along with a placebo in 4-10 month-old infants, 17-19 per group. After analyzing seroresponses to the vaccine in this group, the lo3 PFU dose was discontinued and the remaining infants received either placebo or lo4 PFU of RRV vaccine. The vaccine was diluted in Similac formula (previously shown to lack rotavirus-neutralizing activity) buffered with bicarbonate. Placebo consisted of buffered formula alone. Feedings were withheld from the infants for a t least 2 hr before and 1-2 h r after vaccine

or placebo was given. Prior to vaccinelplacebo administration the children received 30 ml of buffered formula (Similac containing 400 mg bicarbonate). The vaccine was given orally by using a tuberculin syringe without the needle.

Follow U p In the phase I component of the study, each of the

children was visited at home at least once before and daily for the week following administration of vaccine or placebo. During these visits (carried out initially by physicians and later by nurses in those age groups in which the vaccine had been proven safe), the children were examined, their temperatures were taken, and a questionnaire on their clinical status was completed. A second temperature was obtained and recorded in the evening by the mothers. Daily stool specimens were collected. Antecubital venous blood was obtained from all the children within 72 hr prior to administration of vaccine or placebo and 4-5 weeks after.

For the efficacy trial, home visits (twice a week) were conducted by nurses over a 1 year period. Information was gathered on the occurrence of diarrhea, the number and consistency of the child’s stools, occurrence of vomiting, fever, dehydration, and any other symptom associated with the diarrhea. Diarrheal specimens were collected by the nurses during the visits and the mothers were instructed to save stool specimens if diarrhea were to occur between visits.

For the purpose of analysis, diarrhea was defined as the occurrence of three or more liquid or semiliquid stools, or a single stool with mucus or blood during a 24 h r period. An episode occurring 48 or more hr after the end of a previous diarrheal episode was considered as a new episode.

Most episodes, excepting those of 1 day duration, were seen by the pediatrician, who reviewed and collected any additional clinical information. Treatment was carried out according to World Health Organiza- tion standard procedures. Breastfeeding was promptly resumed and other foods were progressively restarted according to the children’s tolerance.

Laboratory Studies Vaccine shedding. Shedding of rotavirus vaccine in

the stools collected during the week post-vaccination was evaluated by direct testing of a 10% stool suspen- sion in a confirmatory preipost ELISA [Kapikian et al., 19791 and by tissue culture isolation as previously described [Hoshino et al., 19841.

Serological responses to the vaccine. Paired serum specimens obtained before and 4-5 weeks after vaccinaton were tested by complement fixation, neutralization, and IgA ELISA assays. Complement fixation assay was performed as described [Kapikian et al., 19791 by using the animal “0” agent rotavirus as the antigen.

Two types of neutralization assays were performed: 1) plaque reduction employing MA104 cells grown in

Efficacy of RRV in Diarrheal Diseases

six-well plates with 60% reduction in the number of plaques considered significant, and 2) evaluation of inhibition of cytopathic effect (CPE) in tube cultures of MA104 cells, with total inhibition considered as end point. In both assays RRV vaccine was used as the antigen [Hoshino et al., 19841. Serial dilutions of sera were incubated with pretrypsinized virus and then absorbed for 1 hr onto the plates or added directly to the tubes containing MEM. The virusiserum mixture was then washed, and melted agarlMEM (for the plaque neutralization assay) was added. The virus1 serum mixture was not washed in the tube neutralization assay. The test was read 3-5 days later for the plaque reduction neutralization assay and at various intervals up to 7 days later for the tube neutralization assay.

For IgA ELISA, 96-well microtiter plates coated with goat antirotavirus antibody were used to capture rotavirus antigen (RRV). Addition of serum dilutions to the wells followed by an alkaline phosphatase-anti-human IgA conjugate allowed titration of rotavirus specific anti-human IgA in the serum [Losonsky et al., 19881.

Detection of wild-type rotaviruses. Suspensions (10%) of the stool specimens isolated during the follow- up period were tested for rotavirus by the confirmatory ELISA [Kapikian et al., 19791 performed initially in the field and later reconfirmed (98% agreement) a t the National Institutes of Health. Positive specimens were subjected to tissue culture isolation as previously described [Hoshino et al., 19841.

Fresh examination of stools. Stools were examined by light microscopy with and without iodine staining within 4 hr of collection to detect parasites or their cysts and helminth ova. In addition, Cryptosporidium was identified by using a modified Ziehl-Neelsen staining [Hendriksen and Pohlenz, 19811, and Entarnoeba histolytica by an immunoperoxidase technique previously described [Perez et al., 19871.

Detection of bacterial enteric pathogens. Fresh stools or rectal swabs transported in Cary Blair medium were examined by conventional techniques for the detection of common enteric pathogens [Lennette et al., 19801. Further characterization was made with the API system. Campylobacter detection was carried out in Campy-Bap medium [Luechtefeld et al., 19811 with 5% human blood; plates were incubated at 42°C in a 5% CO, atmosphere for 48 hr. For Yersinia enterocolitzca, SalmonellaiShigella plates and McConkey plates were incubated for 48 h r at room temperature.

E. coli colonies were further tested to assess the pres- ence of heat-labile (LT) and heat-stable (ST) enterotoxin-coding genes with DNA hybridization probes [Moseley et al., 19821 generously provided by F. Ca- bello. In addition, for ST detection infant mice were inoculated and examined by established procedures [Dean et al., 19721. Enteropathogenic E. coli strains were defined by their reactivity with antisera to 0 and H antigens as detected with commercially available antisera [Levine and Edelman, 19841.

221

Data Analysis The chi-square test, Student's T test, or the Fisher

exact test was employed where appropriate to analyze the data obtained.

Diarrhea incidence was calculated from the morbidity data obtained during the surveillance period. The days for which information was available from the home visits were used to calculate the number of episodes per child per year.

Vaccine efficacy was calculated from the crude rates of rotavirus diarrhea in the placebo and vaccine groups. Its significance was assessed by X2 analysis.

RESULTS Vaccine Safety

Since the lo5 PFU dose of RRV vaccine had previously been shown to be unacceptably reactogenic in 6-12 month-old infants [Vesikari et al., 1986; Gothe- fors e t al., 19891, we did not test this dose in infants less than 6 months of age even though a t this age passively acquired maternal antibody might protect against vaccine reactions. Thus, we evaluated two lower doses of vaccine (lo3 and lo4 PFU) in infants under 6 months of age. The lo4 PFU was adapted as the standard dose since, as shown later, its antigenicity was greater than that of the lower dose. For the field trial the lo4 PFU dose was administered to 151 children, whereas 151 received the placebo preparation. The reactogenicity of this dose of vaccine is shown in Table I.

Although a temperature of 38.1"C or higher occurred more frequently in vaccinees than in controls, this difference was not statistically significant. However, a febrile episode of 2 39°C occurred significantly more often in the vaccinees. Diarrhea, as defined by the pres- ence of three or more liquid or semiliquid stools in any 24 hr period, was observed during the week postinocu- lation in 13 of the 151 vaccinated and in nine of the 151 control infants. The occurrence of diarrhea in the vaccine group was not significantly different from the placebo group. None of the cases of diarrhea was associated with signs of dehydration.

Diarrheal Episodes Detected During the 1 Year Follow Up

A summary of the number of diarrhea cases during the 1 year follow-up period is presented in Table 11. In the 151 children who received placebo 47,267 days of observation were conducted out of a possible maximum of 55,115 days (151 x 365). This corresponds to 129.5 persodyears, or 86% coverage, and includes the days of observation of 28 children who did not complete the study. The corresponding figure for the children who received the lo4 PFU dose of vaccine was 123.5 person/ years (82% coverage, 34 children in this group did not complete the study). In the 18 children who received the lo3 PFU dose of vaccine, surveillance coverage was 94% (17 personiyears).

The average number of diarrheal episodes was sim-


TABLE I. Clinical Reactions to RRV Vaccine During the Week Following. Vaccination

Controls Vaccinees No. of infants studied 151 (75)" 151 (76) No. of infants with fever 2 38.1"C 16 (5) 26 (10) Mean duration of fever (days) 1.4 (1.4) 1.5 (1.5) No. of infants in whom fever

started on day 0 4 (3) 6 (1)

3 3 (1) > 3 2 (0)

0:K (0)

9 (4) 1.8 (1.5)

No. of infants with fever 2 39°C No. of infants with diarrhea

Mean duration of diarrhea (days) No. of infants in whom diarrhea

(23 liquid or semiliquid stoolsiday)

started on day 0 3 ( 2 ) 1 0 (0) 2 0 (0) 3 4 (1)

> 3 2 (1)

Coughing 26 (101 Vomiting 18* (5)* Bronchiolitis l ( 0 )

Other findings

"Data in parentheses corresponds to 1-4 month-old infants. "The fever in four of these infants began the day of or the day after vaccination; three of them developed illnesses (bronchiolitis, otitis, and pharyngitis) not characteristically associated with rotaviruses. .'P ~..05 by chi-square analysis.

ilar in the three groups: 2.2,2.3, and 2.8 episodeslchildl year in the placebo, lo4 PFU vaccine group, and lo3 PFU vaccine group respectively. Age-specific diarrheal rates ranged from 0.13 to 0.20 episodeslchildlmonth in both the vaccine and placebo groups (Table 111).

Etiology of Diarrheal Episodes Of 632 diarrheal episodes recorded, specimens were

available for rotavirus detection (frozen or fresh stools) in 478 cases (77%). Fresh examination to search for parasites was carried out in 372 diarrheal specimens and bacterial isolation and identification (from fresh stool specimens or rectal swabs) in 395 specimens. The episodes for which specimens were not available repre- sent those reported by the mothers after the diarrheal episode had stopped.

The identification rates for the different agents associated with diarrhea according to vaccinelplacebo groups are shown in Table IV. The table also includes the results obtained from etiological studies of control stools obtained randomly from the same children during the surveillance period a t times when they were free of diarrhea for a t least 3 weeks.

Overall, toxin-producing E . coli strains (ETEC) were the most commonly detected agents. Strains producing only heat-stable toxin (ST) were more common than those producing only heat-labile toxin (LT) or strains producing both ST + LT. ST ETEC was detected in 119 of 350 diarrheal episodes studied and in 67 of 295 con-

TABLE 11. Diarrheal Episodes Detected During the 1 Year Follow-Ur, Period

Vaccine Placebo lo4 PFU lo3 PFU All

No. infants studied 151 151 18 320 No. personidays of

roiiow up 47,267 45,086 6,236 98,569 No. personiyears of

follow up 129.5 123.5 17.1 270.1 No. diarrheal episodes 290 294 48 632 Diarrhea rates"

No. infants who did not (episodesichildiyear) 2.2 2.3 2.8 2.3

complete the study (%) 28 (19) 34 (23) l ( 6 ) 63 (20)

"No. episodesiNo. personsiyear of follow-up.

trol non-diarrheal stools (34% vs. 23%, P<.002). En- teropathogenic E. coli (EPEC) was the second most commonly detected agent. It was isolated in 43 of 395 diarrheal stools or rectal swabs and in six of 342 control stools examined (10.9% vs. 1.8%, P<.OOl). Rotaviruses were detected in 33 of 478 diarrheal stools examined (6.9%) but only in one of 298 (0.3%) control stools (P<.OOl). Cryptosporidium was associated with 18 of 372 episodes studies and seen in only one of 275 control stools examined (4.8% vs. 0.4%, P <.001). Campylobac- ter was detected in five cases and Shigella sp. and Sal- monella sp. in two cases each of 395 diarrheal specimens tested; none of these three agents was found in control stools.

Detection of the etiologic agents in diarrheal stools of the vaccine and placebo groups was not significantly different except for rotavirus (compare the third and fifth columns of Table IV). Age-specific rates for the agents detected in the combined placebo and vaccine recipients are shown in Table V. The lowest detection rates for all agents were observed in infants 1-4 months old, except for ST-producing ETEC which was detected in 40% of the cases tested in this age group. Similar rates of ETEC isolation were seen in the other age groups. Most cases of EPEC occurred in the 5-8 and 13-20 month groups while rotavirus occurred mostly in 5-12 month-old infants. Parasites (Cryp- tosporidiurn, E . histolytic, and Giardia) were most commonly observed after 9 months of age.

The severity of diarrhea, as determined by its duration, number of stools per day, and concomitant symp- toms and signs, was analyzed according to the agents detected in Table VI. A modification of the scoring system originally reported by Hjelt et al. [1986] and adopted by us in previous studies [Flores e t al., 19871 was used as a composite index of severity. Rotaviruses were associated with more cases of diarrhea which lasted more than 3 days, which were associated with fever and dehydration and which had higher severity scores than diarrheal episodes associated with any of the other agents; 48% of the rotavirus illnesses had a severity score equal or higher than 7 in comparison to

Efficacy of RRV in Diarrheal Diseases 223

TABLE 111. Age Distribution of Diarrheal Episodes According to VaccinelPlacebo Administration

Age at the time No. episodes detected of episode (months) Plac. Vac. lo4 PFU Vac. lo3 PFU All

in recipients of

1-4 33 38 0 71 5-8 64 70 10 144 9-12 99 78 11 188 13-16 65 78 16 159 17-20 25 27 7 59 21-23 4 3 4 11 All 290 294 48 632

Child' months

of observation

359 951

1,261 1,078

449 62

4.160

~

No. episodes1

childlmonths of

observation 0.20 0.14 0.15 0.15 0.13 0.15

TABLE IV. Agents Detected in Diarrheal and Control Stools During the One-Year Follow-Up

Diarrheal specimens from indicated group Placebo Vaccinea Allb Control specimens

Aaent detected No. tested % No. tested % No. tested % No. tested % No. positivel No. positive No. positivel No. positivel

Rotavirus ETEC (ST) ETEC (LT) ETEC (STiLT) EPEC C. jejuni Shigella sp. Salmonella sp. Cryptosporidium E. histolvtica

221223 511170 121170 71170

231170 21170 11170 01170 71171 61171

9.9 30.0

7.1 4.1

13.5 1.2 0.5 0.0 4.1 3.5

81226 631170 101170 71170

151170 31170 01170 11170 91171 41171

G. larnbiia 131171 7.6 51171

"lo4 PFU dose. bIncludes children receiving placebo, los and lo4 PFU dose of vaccine.

3.5 331478 37.0 1191350

5.9 261340 4.1 161355 8.8 431395 1.7 51395 0.0 21395 1.1 21395 5.2 181372 2.3 101372 2.9 171372

6.9 11298 34.0 671295 7.6 151240 4.5 51232

10.9 61342 1.3 01342 0.5 01342 0.5 01342 4.8 11275 2.7 01275 4.6 71275

0.3 22.7 6.3 2.2 1.8

0 0 0

0.4 0

2.5

TABLE V. Agents Detected According to Age When Diarrheal Episodes Occurred (Includes Data From Placebo and Vaccine Recipients)

~

Months Agents detected 1-4 5-8 9-12 13-16 17-20 21-22 Rotavirus 1160 (2)" 111109 (10) 121139 (9) 71119 (6) 2144 (5) 1/10 (10) ETEC (ST) 23158 (40) 30181 (37) 331105 (31) 19186 (22) 10130 (33) 4110 (40) ETEC (LT) 2/48 (4) 7181 (9) 11195 (12) 5186 (6) 1130 (3) 0110 (0) ETEC (ST + LT) 0148 (0) 3181 (4) 5195 (5) 7186 (8) 1130 (3) 0110 (0)

1/10 (10) EPEC 2150 (4) 10191 (11) 91109 (8) 171101 (17) 4134 (12) Cryptosporidium 0148 (0) 0183 (0) 61107 (6) 10193 (11) 2131 (6) 0110 (0) E. histolytica 0148 (0) 2183 (2) 21107 (2) 4193 (4) 2131 (6) 0110 (0) Giardia 0148 (0) 1183 (1) 51107 (5) 6193 (6) 4/31 (13) 1/10 (10)

"No. positive1No. tested (%)

13.3% of the illnesses in which other agents were detected. Four of the seven cases of dehydration which occurred during the study were associated with rotavirus. No major differences were observed in the clinical features of the illnesses due to other agents except for the association ofE. histolytica with fever and illness of longer duration.

Seroresponses to the Vaccine, Vaccine Shedding In the initial studies in which a lo4 PFU dose of

virus was compared to the lo3 PFU dose, the serore-

sponse rates as studied by either CF or neutralization assays were significantly lower in the children receiving lo3 PFU [Perez-Schael et al., 19873. Table VII shows the seroresponses by CF, neutralization, and IgA ELISA in the lo4 PFU vaccine and placebo groups. Few responses were observed by CF and or neutralization assays in the 1 month-old group but 50% responded by IgA ELISA. Responses were most commonly detected in 2-4 month-old infants; 75-87% of them developed a response by any test. Neutralization assays appeared to be more sensitive for detecting seroresponses than


TABLE VI. Clinical Features of Diarrheal Episodes According to Agent Isolated (Includes Data From Placebo and Vaccine Recipients)

Mean Agent No. of duration lasting stools/ 2 4 stools/ severity detected cases (days) 23 days day day Vomiting Fever Dehydration scorea Rotavirus 33 3.6 18 (55) 6.6 25 (76) 8 (24) 18 (55) 4i12) 6.5 ETEC (ST) 119 3.3 25 (21) 4.9 60 (50) 33 (28) 24 (20) l(1) 4.7 ETEC (LT) 26 2.7 8 (31) 6.6 21 (81) 9 (35) 6 (23) 0 (0) 4.7 ETEC (STILT) 16 4.7 6 (38) 4.6 7 (44) 5 (31) 5 (31) 0 (0) 4.8 EPEC 43 3.2 16 (37) 4.4 20(46) 12 (28) 9(21) 1(2) 5.0 C. jejuni 5 2.2 0 (0) 4.8 3 (60) 2 (40) 0 (0) 0 (0 ) 4.2 Cryptosporidium 18 3.5 7 (38) 3.5 13 (72) 8 (44) 3 (17) l ( 6 ) 5.4 E . histolytica 10 6.0 4 (40) 6.0 9 (90) 3 (30) 5 (50) 0 (0) 5.3 G. lamblia 17 3.9 7 (41) 4.3 11 (64) 0 (0) 5 (29) 0 (0) 4.7

Mean No. (%) MeanNo. No. (%) with

"Severity score as determined by a modification of that proposed by Hjelt et al. 119861.

No. ('%I with score

2 7 16 (48) 16 (13) 3 (12) 2 (12) 6 (14) 0 (0) 3 (17) l ( 1 0 ) 3 (17)

TABLE VII. Seroresponses to Rotavirus After Administration of RRV Vaccine ( lo4 PFU) or Placebo

Comple- Assav ment Neutral- ELISA emDfoved fixation izaton IeA Anv Age, inoculum

1 month Placebo 1/10 (10)" Vaccine 3111 (27)

Placebo 1/19 (5) Vaccine 6114 (43)

Placebo 3/21 (14) Vaccine 11/25 (44)

Placebo 3/18 (17) Vaccine 7/15 (47)

Placebo 8/68 (12) Vaccine 17/45 (62)

Placebo 12/73 (16) Vaccine 35/69 (51)

Placebo 201141 (14) Vaccine 661134 (49)

2 months

3 months

4 months

1-4 months

5-10 months

1-10 months

0/3 (0) 1/10 (10) 2/11 (18) 218 (25) 5/10 (50) 5/11 (45)

1118 (13) 1/16 (6) 2/20 (10) 7/11 (64) 11/14 (79) 12/16 (75)

2/12 (17) 3/22 (14) 4/22 (18) 13/18 (82) 18/26 (69) 23/26 (88)

3/13 (23) 6/18 (33) 6/18 (33) 10/14 (71) 7/13 (54) 12115 (80)

6/36 (17) 11/66 (17) 14/71 (20) 32/51 (61) 41/63 (65) 52/68 (76)

11/35 (31) 9/71 (13) 18/73 (25) 24/47 (51) 31/69 (45) 47/78 (60)

17/71 (24) 201137 (15) 321144 (22) 56/98 (57) 721132 (55) 991146 (68)

'No. of children with a 24-fold seroresponseiNo. tested (%)

the CF assay, whereas neutralization and IgA ELISA were comparable. By ELISA IgA testing, responses were detected in 15% of the children who received placebo and 55% of the vaccinees for the entire group of 1-10 month-old infants. Overall, 22% of the children who received placebo and 68% of the vaccinees developed a seroresponse by any of the tests employed.

Shedding of vaccine was detected in 64% of the children receiving the higher dose tested. This was significantly higher (P < .05) than vaccine shedding observed in children receiving lo3 PFU (29%). Vaccine shedding could only be detected by tissue culture isolation in most instances.

Rotavirus Diarrhea During the follow-up year, 22 rotavirus-positive

cases (9.9%) were detected among 223 diarrheal

episodes which occurred in children who had received placebo. This corresponds to an incidence of 0.169 episodes of rotavirus disease per childiyear; in contrast, eight of 226 (3.5%) diarrheal specimens from the children who previously received the lo4 PFU dose of RRV vaccine yielded rotavirus (0.064 rotavirus episodes/child/year). Specimens were not available for rotavirus testing from 67 diarrheal episodes involving control infants and 58 episodes occurring in vaccine recipients. Rotavirus was detected in three of 32 (9.4%) diarrheal episodes occurring in the 18 infants who had received the 10" PFU dose of vaccine (0.175 episodes/ childiyear).

The clinical features of rotavirus episodes are sum- marized in Table VIII. The difference in the occurrence of rotavirus diarrhea between control and vaccine groups corresponds to an overall efficacy of 64% (P <.01). While the severity of illness appeared to be milder in the infants receiving vaccine in comparison to the placebo group as judged by the duration of diarrhea, vomiting, and fever; the frequency of vomiting, fever, and dehydration; and the need for hospitaliza- tion, most of these differences (with the exception of the mean duration of vomiting and number of vomiting episodes per day) were not statistically significant. The mean age of diarrheal illness was significantly greater in vaccinees (P <.05).

Six of the eight vaccinees who developed rotavirus diarrhea during the follow-up period had developed a rise in rotavirus antibody following vaccine administration.

Table IX analyzes the number of rotavirus illnesses in the two treatment groups according to the composite severity score and Table X presents an analysis according to the age at time of vaccination. While overall vaccine efficacy was 64%, higher levels (up to 90%) were observed against episodes of greater severity. Ef- ficacy levels of 80-86% were observed in the group of infants who were 1-4 months of age at the time of vaccination (Table IX, bottom). Efficacy was not observed in infants vaccinated after the fifth month of age; however, only five rotavirus illnesses occurred in these infants (Table X).


TABLE VIII. Clinical Features of Rotavirus Diarrhea in Children Who Received RRV Vaccine (lo* PFU) or Placebo

225

Placebo Vaccine (151) (151) P

No. of rotavirus illnesses 22 8 <.01 Mean duration of diarrhea

(days k SEM) 4.0 t 0.4 2.8 2 0.6 ns Mean no. of stoolsiday

-+ SEM 6.2 2 0.5 6.8 +- 1.3 ns No. with vomiting 14 7 Mean duration of vomiting

Mean no. of vomiting

No. of cases with fever 14 5 Mean duration of fever

(days t SEM) 2.3 ? 0.4 1.6 k 0.2 ns No. of cases with

dehydration 3 1 ns No. of cases requiring

hospital admission 4 1 ns Mean age when diarrhea

occurred (months 5 SEMI" 9.5 +- 0.8 12.8 -+ 1.0 <.05 Mean duration (mo.) from

time of vaccination to diarrhea occurrence 5.9 7.6 ns

Mean severity score (+ SEMI 7.2 ? 0.6 6.5 k 0.9 ns

"The mean age a t the time of vaccination was similar (4 9 months) for placebo and vaccine groups.

(days 2 SEMI 2.9 L 0.4 1 3 t 0.2 <.01

episodesiday 1.86 1.12 <.05

TABLE IX. Protective Efficacy of RRV Vaccine Against Rotavirus Diarrhea According to Severity of

Diarrheal Illness

All infants (1-10 months old) No. patients with

indicated score Severity Controls Vaccinees 76

Any (3-13) 22 8 64 < . O l score (n=151) (n=151) efficacy P*

6 15 4 73 <.01 7 13 3 77 <.05 8 10 1 90 c.01 9 7 1 85 <.05

1-4 month-old infants No. patients with

indicated score Severity Controls Vaccines % score (n = 75) (n=76) efficacv P* Any (3-13) 17 3 83 <.001

6 12 2 83 <.005 7 10 2 80 <.05 8 7 1 86 <.05 9 ti 1 84 .051

*Chi-square analysis.

The serotypes of the rotavirus strains detected are shown in Table XI. Ten of the 19 specimens from placebo recipients and three of six specimens from vaccinees which could be serotyped belonged to serotype 3; protection against serotype 3 rotavirus was statistically significant (P <.05). Serotype 2 was the second

TABLE X. Distribution of Rotavirus Diarrhea According to Age at the Time of VaccineiPlacebo Administration

Placebo Vaccine Age No. with rotavirus No. with rotavirus (months) No. diarrhea No. diarrhea 1 11 2 22 3 23 4 19 5 17 6 14 7 13 8 12 9 14

10 6 Total 151

13 18 28 17 20 11 11 12 10 11

151

2 (1) 0 1 0 1 0 3 0 0 1 8 (1)

"The number of infants with an illness severity score of 8 or more is shown in parentheses.

TABLE XI. Serotype Specificity of Wild-Type Rotaviruses Associated With Diarrheal Illness in Children Who

Received RRV Vaccine or Placebo

Placebo Vaccine (n=151) ( n = 151)

No. of cases of rotavirus diarrhea 22 8 Serotype 1 3 0 Serotype 2 5 3 Serotype 3 10" 3 Serotype 4 1 0 Not typed 3 2

'FP <.05 (chi-square analysis).

most prevalent type, being detected in five control infants and three vaccinees. Only three serotype 1 and one serotype 4 strains were identified, all of them in the control group. The severity of diarrhea was similar for the four serotypes and ranged between 6.3 and 7.9 using the scoring system employed.

Other etiologic agents were present in 12 of the 33 stool specimens in which rotavirus was detected. They include ETEC in nine cases (seven in the control and two in the vaccine group), EPEC in three cases (all in control children), and Campylobacter and Crypto- sporidium in one case each in vaccine and placebo recipients respectively.

Relative Severity of Diarrheal Episodes According to Etiology and

Vaccine/Placebo Administration Of 15 diarrheal episodes with a score of 8 or more in

the placebo group, 10 (67%) were due to rotavirus, whereas only one of seven (14%) with such a severity score in the vaccine group was rotavirus positive (Table XII). Etiological studies on the non-rotavirus cases with diarrheal severity of 8 or greater showed the pres- ence of ETEC in five cases, EPEC in two cases, and Cryptosporidium in one case. In addition, of the 11 rotavirus cases, three were accompanied by ETEC, two by EPEC, and one by Cryptosporidium.


studies carried out in Bangladesh, Brazil, and Mexico respectively in which incidence figures of up to 10 episodes per child per year were detected. Our lower rates may be due to the fact that despite the poor living conditions in Los Erasos and Cotiza, water and sewage disposal are available, all the enrolled mothers had a t least a primary level of education, and most children are under pediatric surveillance and receive conventional immunizations. The incidence of diarrhea in this study is comparable to the estimated 2 episodesichildi year observed by Guerrant et al. 119831 in a subsample of children in whose households piped water and flush toilets were available. The incidence of diarrheal episodes did not appear to vary significantly in relation with the children’s age (range 0.13-0.20 episodes per childimonth).

Despite application of the established bacteriological and parasitological methods of detection, an agent was not detected in a large proportion of cases, and the contribution of some of them was difficult to evaluate since they were commonly found in control specimens obtained from the same children during diarrhea-free periods. The most common agent detected from the diarrheal episodes studied was heat-stable toxin (ST)- producing E. coli, which was identified in 34% of the diarrheal specimens examined; strains producing both heat-labile toxin (LT) and ST were detected in 4.5% of specimens while LT-only-producing E . coli was detected in 7.6% of the specimens tested. These results are similar to previous longitudinal studies carried out in Bangladesh [Black et al., 19801 and Brazil [Guerrant e t al., 19831 in which toxin-producing E . coli strains were the most commonly detected agents in diarrheal specimens. Toxin-producing E. coli strains were frequently isolated from control specimens as well; thus, ST-, LT-, and LT + ST-producing E. coli strains were seen in 22.7%, 6.310, and 2.2% of control specimens tested, respectively. Although the detection rates of ST and ST + LT were significantly higher in diarrheal than in control specimens, those for LT ETEC were not. These results contrast with observations in Bang- ladesh where ETEC strains were only found in 1-210 of control specimens. On the other hand, examination of control specimens from several cross-sectional studies in children admitted to hospitals has also indicated high rates of isolation of ETEC [Levine, 19871, high- lighting the difficulty in assigning an etiological role to toxin-producing E. coli in individual cases.

Enteropathogenic E . coli strains (EPEC) were detected in 10.9% of the diarrheal and in 1.8% of the control specimens examined. These detection rates are comparable to those described in several cross- sectional studies of diarrhea [Levine and Edelman, 19841 and in the home-based study in Brazil [Guerrant e t al., 19831.

The detection rates for other common diarrhea-producing bacteria, Campylobacter jejuni, Shigella, and Salmonella, ranged from 0.5 to 1.3%. Isolation rates for Shigella in other studies have been higher [Black et al.,

TABLE XII. Rotavirus Diarrheal Episodes of Greater Severity According to VaccineiPlacebo Administration

Rotavirus Rotavirus ‘% of cases due positive negative All to rotavirus

Episodes with a severity score 2 8 Vaccine group 1 6 % 7 14 Placebo group 10 5 15 67 All 11 11 22 48

Episodes with a severity score 2 7 Vaccine group 3 15” 18 17

All 16 35 51 31 Placebo group 13 20 33 39

,’Excludes a confusing case of an infant who presented five episodes of non-rotavirus diarrhea during a 2.5 month period.

Thirteen of 33 (39%) of the episodes with a severity score of 7 or greater among the control children had rotavirus vs. three of 18 (17%) in the vaccinees.

DISCUSSION The goal of preventing diarrheal diseases in infants

and young children requires knowledge of the natural history of infection produced by the different agents responsible for this syndrome. While many laboratory studies have been carried out to characterize these agents, few prospective field studies have been performed to assess their relative contribution and to pre- dict the impact that specific vaccines might have in diminishing diarrheal morbidity and mortality. In the present study we investigated the etiology of diarrheal disease in 320 infants in two poor neighborhoods of Caracas a t the time that a rotavirus vaccine candidate, the rhesus rotavirus vaccine MMU-18006, was evaluated for efficacy. The children were under intensive surveillance (home visits) for 1 year after receiving the vaccine or a placebo in order to detect and grade diarrheal episodes, and to collect specimens for etiological studies.

Although the RRV vaccine proved to be efficacious, there was no difference in the overall incidence of diarrhea observed between the children receiving it and a placebo group. This is a reflection the relatively low incidence of rotavirus diarrhea in longitudinal home- based studies and predicts a low impact of the vaccine on the overall incidence of diarrhea as such. On the other hand, rotavirus was commonly associated with the more severe cases, and the vaccine was effective against this type of diarrhea. This observation needs t o be expanded in future, larger studies since only seven cases of severe dehydrating diarrhea were detected in this trial.

The overall diarrheal rates determined in this study (2.3 episodesichildiyear) are lower than those estimated by Snyder and Merson, who reviewed several studies conducted before 1982 in countries of Asia, Af- rica, and Latin America [Snyder and Merson, 19821, and those reported by Black et al., /19801, Guerrant et al. 119831, and Sepulveda et al., 119881 in more recent


1980; Guerrant et al., 19831, but this agent has not been commonly found in previous cross-sectional studies in Caracas. Likewise, a higher incidence of Campy- Zobacter in longitudinal studies in Mexico and in Brazil has been reported [Sepulveda et al., 1988; Guerrant et al., 19831. This agent was previously detected in 10% of the cases in a cross-sectional study in Venezuela [Ur- restarazu et al., 19871.

Cryptosporidiun was found in 4.8% of the diarrheal specimens but only in one control stool of 275 tested, whereas E. histolytica was detected in 2.7% of the diarrheal specimens and none of the control specimens. Cryptosporidia had been detected in previous cross-sectional studies in Venezuela [Perez-Schael et al., 19851. Giardia lamblia was detected in 4.6% of diarrheal specimens vs. 2.5% of control stools.

Rotaviruses were identified in 22 of 223 (9.9%) of the diarrheal episodes of children in the placebo group in comparison to eight of 226 (3.5%) diarrheal episodes of vaccinees. Only one control stool tested positive for rotavirus. All four serotypes of rotavirus were identified, but serotype 3 was the most common (13 cf 25 specimens that could be typed were serotype 3), followed by serotype 2. The number of serotype 3 rotavirus strains identified in the control group was significantly higher than in the vaccinees (P <.05).

Illnesses produced by rotavirus appeared to be more severe than those produced by other agents in terms of duration and occurrence of fever. When a severity score was employed to rate each diarrheal episode, about half of the rotavirus episodes scored over 7 whereas one in six to one in 10 of episodes associated with the other agents examined had such severity scores. Likewise rotaviruses were associated with four of the seven illnesses in which dehydration accompanied the diarrheal syndrome. Similar observations have been made in other studies: for example Wyatt et al. [1979], who followed a cohort of 24 Guatemalan children for 3 years, determined that half of the diarrheal episodes that led to dehydration were due to rotavirus. Like- wise, in the study by Black et al. [1980] in Bangladesh, 39% of the episodes in which dehydration occurred were associated with rotaviruses; this frequency was significantly greater than corresponding frequencies observed in illnesses associated with ETEC, ShigelZa, or illnesses in which a pathogen was not detected.

The current goal for a rotavirus vaccine is to protect against severe cases of rotavirus gastroenteritis that can lead to dehydration and death. Several studies suggest that natural rotavirus infecton does not prevent subsequent illness, but rather modifies later exposure since second or third infections appear to have dimin- ished severity [Chiba et al., 1986; Bishop et al., 19831. Repeated rotavirus infections may boost and perhaps broaden the primary antibody response [Green et al., 19891. For instance, Bishop et al. [1983], who followed for up to 3 years a group of children who were infected with rotavirus during the neonatal period and a group of children who were not infected observed that subse-

227

quent rotavirus infection and diarrhea occurred with almost equal frequency in both groups; however, none of the children in the neonatally-infected group suf- fered severe disease, whereas many cases of severe dehydrating diarrhea were seen in the control group. Al- though the rotavirus strains responsible for diarrheal illness were not serotyped in that study, the neonatally acquired strain belonged to serotype 3. It is likely that during the 3 year observation period rotavirus serotypes other than type 3 caused infection in both groups, suggesting that heterotypic protection against severe illness was provided by a combination of the initial and subsequent infections. Studies with the RIT 4237 strain, a serotype 6 bovine rotavirus vaccine candidate, carried out in Finland showed that heterotypic protection could be induced in children vaccinated after the sixth month of life [Vesikari et al., 19841. Similar studies in Philadelphia with another serotype 6 bovine rotavirus strain, the WC3 vaccine candidate, also showed significant protection against rotavirus diarrhea [Clark et al., 19881. In these two field trials the rotavirus strains circulating at the time of the follow-up period were predominantly serotyped as 1, indicating that these vaccines induced heterotypic protection. Be- cause of the age of the children vaccinated, it is likely that they had been previously infected with rotavirus, since subsequent studies in younger infants who were immunized with the RIT 4237 vaccine in Rwanda and Gambia failed to demonstrate vaccine efficacy. Exploit- ing the concept of infection-permissive immunity [Kil- bourne, 19841 may be relevant to rotavirus vaccinology given the relatively large number of human rotavirus serotypes (six up to date) [Wyatt et al., 1982; Matsuno et al., 1985; Clark et al., 19871.

In the present study, the RRV vaccine, which is similar to human serotype 3 rotavirus (through its VP7 neutralization protein), induced a high level of protection against severe episodes of diarrhea, especially when it was fed to the younger infants in the group (1-4 months old). As noted above, most of the rotavirus illnesses occurring during the study period were due to serotype 3 rotavirus which was prevalent in Caracas a t the time. It was of interest that serotype 2 rotavirus illness was observed in three vaccinees and five controls, suggesting in these small numbers that protection against that serotype was not provided.

Other studies with RRV vaccine have failed to detect significant protection. However, during these studies, in which young infants (2-5 months old) were immunized, serotypes other than serotype 3 strains were circulating in the community. For instance in a field trial in Rochester [Christy et al., 19881 most of the viruses isolated were serotype 1, whereas in a separate study in an Indian reservation in Arizona [Kapikian et al., 1988133 only one of the strains isolated was a serotype 3. This failure to induce heterotypic protection contrasts with the studies with the RIT and WC3 vaccines described above, and perhaps can be explained on the


basis of prior rotavirus infection of the infants (of older age groups) who received these two vaccines.

The present study shows that administration of RRV vaccine to the younger infants resulted in a high level of protection (up t o 86% efficacy). Although febrile episodes occurred in association with vaccine administration, the reactions were moderate and acceptable. Di- arrhea or other clinical findings were not associated with vaccination.

Our present study, analyzed in combination with the studies in Rochester and Arizona, indicates that the protection afforded by RRV vaccine in young infants is only homotypic and therefore suggests the need for a polyvalent rotavirus vaccine which includes a t least the four epidemiologically important serotypes. Since rotavirus illness occurs in children of developing countries even a t very young ages (from the second month of life on), there is a strong need to administer rotavirus vaccines as early in life as possible. The results with the RRV vaccine in this study suggest that this may be feasible.

ACKNOWLEDGMENTS We thank the families from Los Erasos and Cotiza

who participated in this study for their enthusiastic support and patience. We thank the following individ- uals for their excellent technical support: Harvey James, Jr., Annie Pittman, Johnna Sears, Cidalia Ur- bina, and Jordi Boher.

This study was supported by funds from the Agency for the International Development (AID) Vaccine Pro- gram through a PASA agreement with the National Institute of Allergy and Infectious Diseases.

REFERENCES Anderson EL. Belshe RB, Bartram J, Crookshanks-Newman F, Cha-

nock RM, Kapikian AZ (1986): Evaluation of rhesus rotavirus vaccine (MMU 180061 in infants and young children. Journal of In- fectious Diseases 153:823-831.

Bishop RF, Barnes GL, Cipriani E, Liund JS (1983): Clinical immunity after neonatal rotavirus infection: A prospective longitudinal study in young children. New England Journal of Medicine 309: 72-76.

Black RE, Merson MH, Rahman ASMM, Yunus M, Abdul Alim ARM, Huq 1, Yolken RH, Curlin GT (1980): A two-year study of bacterial, viral, and parasitic agents associated with diarrhea in rural Bangladesh. Journal of Infectious Diseases 142660-664,

Chiba S, Tokoyama T, Nakata S, Morita Y, Urasawa T, Taniguchi K, LJrasawa S, Nakao T (19861: Protective effect of naturally acquired homotypic and heterotypic rotavirus antibodies. Lancet 2: 417-42 1.

Christy C, Madore HP, Treanor JJ, Pray K, Kapikian AZ, Chanock RM, Dolin R (19861: Safety and immunogenicity of live attenuated rhesus monkey rotavirus vaccine. Journal of Infectious Diseases 154:1045-1047.

Christy C, Madore HP, Pichichero ME, Gala C, Pancus P, Vosefski D, Hoshino Y, Kapikian AZ, Dolin R, Elmwood and Panorama Pedi- atric Groups (1988): Field trial of rhesus rotavirus vaccine in infants. Pediatric Infectious Disease Journal 7:645-650.

Clark HF, Hoshino Y, Bell LM, Groff J , Hess G, Bachman P, Offit PA i 1987): Rotavirus isolate W161 representing a presumptive new human serotype. Journal of Clinical Microbiology 25:1757-1762.

Clark HF, Borian FE, Bell LM, Modesto K, Gouvea V, Plotkin S (1988): Protective effect of WC3 vaccine against rotavirus diarrhea in infants during a predominantly serotype 1 rotavirus sea- son. Journal of Infectious Diseases 158:570-587.

De Zoysa I, Feachem RG (1985): Interventions for the control of diar- rhoeal diseases among young children: Rotavirus and cholera immunization. Bulletin of the World Health Organization 63:569- 583.

Dean AG, Ching Y-C, Williams RG, Harden LB (1972): Test for Es- cherichia coli enterotoxin using infant mice: Application in a study of diarrhea in children in Honolulu. Journal of Infectious Diseases 125:407-411.

Flores J, Perez-Schael I, Gonzalez M, Garcia D, Perez M, Daoud N, Cunto W, Chanock RM, Kapikian AZ (1987): Protection against severe rotavirus diarrhoea by rhesus rotavirus vaccine in Vene- zuelan children. Lancet i:882-884.

Flores J , Kapikian AZ (1988): Rotavirus vaccines. In Farthing MJG, Keusch GT (edsl: “Enteric Infection: Mechanisms, Manifestations and Management.” London: Chapman and Hall, Ltd., pp 509-526.

Flores J, Taniguchi K, Green K, Perez-Schael I, Garcia D, Sears J, Urasawa S, Kapikian AZ (1988): Relative frequency of rotavirus serotype 1,2,3 and 4 in Venezuelan infants with gastroenteritis. Journal of Clinical Microbiology 26:2092-2095.

Green K, Taniguchi K, Mackow E, Kapikian AZ (1989): Homotypic and heterotypic epitope-specific seroresponse in adult and infant rotavirus vaccines: Implications for vaccine development. Journal of Infectious Diseases (in press).

Gothefors L, Wadell G, Juto P, Taniguchi K, Kapikian AZ, Glass RI (1989): Prolonged efficacy of rhesus rotavirus vaccine in Swedish children. Journal of Infectious Diseases 159:753-757.

Guerrant TL, Krichoff LV, Shields DS, Nations MK, Leslie J, deSousa MA, Araujo JG, Correia LL, Sauer KT, McClelland KE, Trow- bridge FL, Hughes JM (1983): Prospective study of diarrheal illnesses in Northeastern Brazil: Pattern of disease, nutritional impact, etiologies and risk factors. Journal of Infectious Diseases 148:986-997.

Henricksen SA, Pohlenz JFL (19811: Staining of cryptospondia by a modified Ziehl-Neelsen technique. Acta Veterinaria Scandinavica 22594-596.

Hjelt K, Grauballe PC, Andersen L, Schiotz PO, Howitz P, Krasilni- koff PA (1986): Antibody response in serum and intestine in children up to six months after a naturally acquired rotavirus gastroenteritis. Journal of Pediatric Gastroenterology and Nutrition 5: 74-80.

Hoshino Y, Wyatt RG, Greenberg HB, Flores J , Kapikian AZ (19841: Serotypic similarity and diversity of rotaviruses of mammalian and avian origin as studied by plaque reduction neutralization. Journal of Infectious Diseases 149:694-702.

Kapikian AZ, Yolken RH, Greenberg HB, Wyatt RG, Kalica AR, Cha- nock RM, Kim HW (1979): Gastroenteritis viruses. In Lennette EH, Schmidt NJ (edl: “Diagnostic Procedures for Viral, Rickettsia1 and Chlamydia1 Infections, 5th ed.” Washington, D.C.: American Public Health Association, pp 927-995.

Kapikian AZ, Flores J, Hoshino Y, Glass RI, Midthun K, Gorziglia M, Chanock RM (1986): Rotavirus: The major etiologic agent of severe infantile gastroenteritis may be controllable by a “Jenne- rian” approach to vaccination. Journal of Infectious Diseases 153: 815-822.

Kapikian AZ, Flores J , Green K, Hoshino Y, Gorziglia M, Nishikawa K, Chanock RM, Perez-Schael I (1988a): Prospects for the development of a rotavirus vaccine against rotavirus diarrhoea by a “Jennerian” and a modified “Jennerian” strategy. In Norrby, SR, Mills J, Norrby E, Whitton W (eds): “Frontiers in Infectious Dis- eases.” NY: Gardiner Coldwell Communications, Ltd, pp 217-232.

Kapikian AZ, Flores J, Midthun K, Hoshino Y, Green KY, Gorziglia M, Taniguchi K, Nishikawa K, Chanock RM, Potash L, Perez- Schael I, Dolin R, Chrysty C, Santosham M, Halsey NA, Clements ML, Sears SD, Black RE, Levine MM, Losonsky G, Rennels M, Gottefors L, Wadell G, Glass RI, Vesikari T, Anderson EL, Belshe RB, Wright PF, Urasawa S (198813): Development of a rotavirus vaccine by a “Jennerian” and a modified “Jennerian” approach. In Ginsberg H, Brown F, Lerner RA, Chanock RM (ed): “Vaccines 88.” Cold Spring Harbor, New York: Cold Spring Harbor Labora- tory, pp 151-159.

Kilbourne ED (1984): Immunization strategy: Infection-permissive vaccines for the modulation of infection. In Chanock RM, Lerner RA (eds): “Modern Approaches to Vaccines.” Cold Spring Harbor, New York: Cold Spring Harbor Laboratory, pp 269-273.

Lennette EH, Balono A, Hausler WJ, J r , Truarit J P (eds) (1980): “Manual of Clinical Microbiology. 3rd ed.” Washington, D.C.: American Society for Microbiology.


Levine MM (1987): Escherichia coli that cause diarrhea: Enterotoxi- genic, enteropathogenic, enteroinvasive, enterohemorrhagic, and enteroadherent. Journal of Infectious Diseases 155:377-389.

Levine MM, Edelman R (1984): Enteropathogenic Escherichia coli of classic serotypes associated with infant diarrhea: Epidemiology and pathogenesis. Epidemiologic Reviews 6:31-51.

Losonsky GA, Rennels MB, Kapikian AZ, Midthun K, Ferra PJ, For- tier DN, Hoffman KM, Baig A, Levine MM (1986): Safety, infec- tivity, transmissibility and immunogenicity of rhesus rotavirus vaccine (MMU 18006) in infants. Pediatric Infectious Disease 5: 25-29.

Losonsky GA, Rennels MB, Lem Y, Krall G, Kapikian AZ, Levine MM (1988): Systemic and mucosal immune responses to rhesus rotavirus vaccine MMU 18006. Pediatric Infectious Disease Journal 7: 388-393.

Luechtefeld NW, Wang WLL, Blaser MJ, Reller LB (1981): Campy- lobacter fetus subsp. jejuni: Background and laboratory diagnosis. Laboratory Medicine 12:481-487.

Matsumo S, Hasekawa A, Mukoyama A, Inouye S (1985): A candidate for a new serotype of human rotavirus. Journal of Virology 54 623-624.

Moseley SL, Echeverria P, Seriwatana J, Tirapat C, Chaicumpa W, Sakuldaipeara T, Falkow S (1982): Identification of enterotoxi- genic Escherichza coli by colony hybridization using three gene probes, Journal of Infectious Diseases 145:863-869.

Perez ES, Perez-Schael I, Perozo-Ruggeri G, Davila D, Romer H, Tapia FJ (1987): Immunocytochemical detection of Entamoeba histolytica. Transactions of the Royal Society of Tropical Medicine and Hygiene 81:624-626.

Perez-Schael I, Boher Y, Mata L., Perez M, Tapia FJ (1985): Cryptosporidiosis in Venezuelan children with acute diarrhea. American Journal of Tropical Medicine and Hygiene. 34:721-722.

Perez-Schael I, Gonzalez M, Daoud N, Perez M, Soto I, Garcia D, Daoud G, Kapikian AZ, Flores J (1987): Reactogenicity and anti-

229

genicity of the rhesus rotavirus vaccine in Venezuelan children. Journal of Infectious Diseases 155:334-338.

Sepulveda J, Willett W, Munoz A (1988): Malnutrition and diarrhea. A longitudinal study among urban Mexican children. American Journal of Epidemiology 127:365-376.

Snyder JD, Merson MH (1982): The magnitude of the global problem of acute diarrhoea1 disease: A review of active surveillance data. Bulletin of the World Health Organization 60:605-613.

Taniguchi K, Urasawa T, Morita Y, Greenberg HB, Urasawa S (1987): Direct serotyping of human rotavirus in stools by an exzyme- linked immunosorbent assay using serotype 1-2-3, and 4-specific monoclonal antibodies to VP7. Journal of Infectious Diseases 155: 1159-1 166.

Urrestarazu MI, Darricarrere RT, Perez M, Daoud G, Serrano N, Cavazza ME, Perez-Schael I (1987): Frequency of Campylobacter jejuni and other agents of acute diarrhea in Venezuelan children. PAHO Bulletin 21240-249.

Vesikari T, Isolauri E, D'Hondt E, Delem A, Andre FE (1984): Pro- tection of infants against rotavirus diarrhoea by RIT 4237 attenuated bovine rotavirus strain vaccine. Lancet i:977-981.

Vesikari T, Kapikian AZ, Delem A, Zissis G (1986): A comparative trial of rhesus monkey (RRV-1) and bovine (RIT 4237) oral rotavirus vaccines in young children. Journal of Infectious Diseases 153332-839.

Wyatt RG, Yolken RH, Urrutia JJ, Mata L, Greenberg HB, Chanock RM, Kapikian A2 (1979): Diarrhea associated with rotavirus in rural Guatemala: A longitudinal study of 24 infants and young children. American Journal of Tropical Medicine and Hygiene 28: 325-328.

Wyatt RG, Greenberg HB, James WD, Pittman AL, Kalica AR, Flores J , Chanock RM, Kapikian AZ (1982): Definition of human rotavirus serotypes by plaque reduction assay. Infection and Immunity 37:llO-115.

Prospective study of diarrheal diseases in Venezuelan children to evaluate the efficacy of rhesus rotavirus vaccine

Documents