Measles causes high morbidity and mortality among young children in countries where measles immunization is not fully implemented.1, 2 In developing countries epidemics can involve 8 to 10% of children in a nonimmunized community,3, 4 with case fatality rates ranging from 5 to 40%.4, 5 From 1985 to 1987 in a population-based study in a depressed urban community in the Philippines, mortality rates from measles were 8.6/1000 in infants<1 year old and 4.8/1000 in children 1 to 4 years.6
Pneumonia complicates measles in as many as 80% of all cases2, 7-10 and is responsible for 25 to 100% of measles deaths in developing countries. Measles is responsible for 6 to 21% of all cases of acute lower respiratory infections (ALRI) and 8 to 50% of all pneumonia-associated deaths in developing countries.2 Among Filipino children admitted into the hospital with ALRI in the mid 1980s, 12.5 to 48% had associated measles and their case fatality rate was 23.5%.11, 12
Pneumonia occurring early in the course of measles is usually caused by the measles virus.9 Superinfection with bacteria or other viruses may develop when the initial symptoms of measles have subsided.2, 13 Bacterial culture of samples obtained by lung puncture has been positive in 55 to 68% of patients with measles,14, 15 but there are few data available on the significance of other viral infections occurring together with measles.12-14, 16
Single dose measles vaccination was introduced into the Philippine Expanded Program of Immunization in July, 1982. The present study was undertaken to investigate the role of measles in ALRI several years after the start of the immunization program, to describe the clinical features of patients admitted for measles-associated pneumonia and to determine its mixed etiology. In addition to clinical diagnosis, a sensitive enzyme immunoassay (EIA) for measles antibody was used, allowing the detection of cases with an atypical clinical presentation.
Patients. All children <5 years old who consulted from March 15, 1989, to August 20, 1990, at the Research Institute for Tropical Medicine in Alabang, a periurban area of Manila, were prospectively enrolled into the study if they had cough for <3 weeks or difficulty in breathing and any one of the following: (1) tachypnea (respiratory rate ≥ 50/min);(2) chest indrawing; (3) inability to drink; (4) cyanosis; or (5) infiltrate on chest roentgenograph even in the absence of the previously enumerated symptoms. The inclusion criteria as well as grading into pneumonia (cough/difficult breathing and tachypnea), severe pneumonia (cough/difficult breathing and chest indrawing) or very severe pneumonia(cough/difficult breathing and cyanosis or inability to drink or other danger signs) were based on the protocol for defining pneumonia recommended at that time by the World Health Organization.17
The clinical diagnosis of measles was based on the typical association of fever, respiratory symptoms and maculopapular rash which subsequently desquamated. Patients with significant measles antibody titers without the clinical diagnosis were also included in the analysis.
In addition to the routine initial radiographic interpretation, chest radiographs were subjected to a blind structured reading by a pediatric radiologist. A dense infiltrate was defined as an infiltrate involving more than one bronchopulmonary segment or an infiltrate≥5 mm in diameter. Most of the hospital's patients came from the lower socioeconomic groups and were walk-in patients, with a small proportion of referrals from other hospitals.
The institutional and ethical review board of the Research Institute for Tropical Medicine approved the study design. Informed consent was obtained from the guardians before enrollment.
Samples. Blood was drawn from all patients for two blood cultures before antibiotic administration. Acute sera, obtained on admission, and convalescent sera, obtained 5 to 21 days later, were stored at −20°C until used for antibody assays. Nasopharyngeal aspirate was collected on admission for virus isolation and viral antigen detection with a pediatric mucus extractor.
Blood culture. From 1 to 2 ml of blood were obtained aseptically from two sites and inoculated into 20 ml of locally prepared blood culture broth containing trypticase soy broth supplemented with sodium polyanethol sulfonate. The blood culture bottles were examined daily for up to 10 days and subcultured after 24 h and 3, 5, 7 and 10 days of incubation regardless of the appearance of turbidity in the blood culture bottles. Subcultures were made onto the following media: plain blood agar and 5 μg/ml gentamicin blood agar for the isolation of Streptococcus pneumoniae, 3 μg/ml bacitracin chocolate agar for Haemophilus influenzae and MacConkey agar for Gram-negative organisms. Identification of isolates was carried out by standard methods.18
Virus isolation. Nasopharyngeal aspirate was inoculated onto monolayers of HEp-2, HeLa Ohio, MA 104, Madin-Darby canine kidney and human foreskin fibroblast cells in 0.1-ml aliquots. Microscopic observations on the inoculated cultures were done daily for the first 5 days and every 48 h thereafter up to 14 days to detect the characteristic cytopathic effect. A hemadsorption test was done on Days 2 to 5 and 7 to 10 to detect parainfluenza and influenza viruses. Cell smears were prepared from cultures showing cytopathic effect or hemadsorption reaction. Isolates were identified by the indirect immunofluorescence staining method with the use of type-specific monoclonal antibodies and fluorescein isothiocyanate-conjugated antimouse IgG.
Viral antigen detection. The enzyme immunoassay methods used have been previously described.19-21
Measles antibody test. Measles IgG and IgM antibody assays were performed at the Department of Virology, University of Turku, Finland, with identical EIA methods as reported previously for influenza A and B virus.22 The measles antigen was prepared as reported earlier.23 A 4-fold rise in either IgG or IgM antibody between acute or convalescent serum or an IgM antibody titer of ≥320 was considered diagnostic.
Other viral antibodies. IgG antibodies against adenovirus, influenza A and B and parainfluenza 1, 2 and 3 were detected by the standard enzyme immunoassay method described by Koskinen et al.22 The antibody titer of the test sera was determined by plotting the mean absorbance reading of duplicate samples on the standard curve and estimating the corresponding EIA units. A 3-fold or greater increase in EIA units between the acute and convalescent sera was considered indicative of a recent infection.
For respiratory syncytial virus antibody determination, the EIA serum dilution method of Meurman et al.24 was used. A 4-fold or greater rise in antibody titer between the acute and convalescent sera was considered diagnostic.
Nutritional status. Standard deviation scores (Z scores) for weight for age, weight for height and height for age were computed based on the National Center for Health Statistics reference population. Children with height for age Z scores of ≤−2 SD were considered stunted, those with weight for height Z scores of ≤−2 SD were considered wasted.25
Statistical methods. Data were analyzed with the SAS software package. Difference in rates between two groups was tested for significance using Fisher's exact test or chi square test with Yates' correction for continuity where appropriate. P < 0.05 was considered significant. Odds ratio was used to determine risk of death among certain subgroups.
Among 2129 patients with any acute respiratory complaint seen at the outpatient department or the emergency room of the hospital during the study period, 350 (16%) children fulfilled the study criteria for ALRI. There were 162 children with typical clinical measles. Of these 99 (61%) were tested serologically for measles antibodies, 95 of whom had a diagnostic measles antibody finding. Four patients with clinical measles had nondiagnostic serology and 3 of these had only 1 serum specimen (Table 1).
Twenty patients had diagnostic measles serology without a clinical diagnosis of measles. However, 10 of these cases had rash by either history or on physical examination, indicating that the diagnosis may have been missed. One had been diagnosed clinically as measles in the same hospital 16 months earlier.
Three children who had received measles vaccination within the previous 3 months and had a diagnostic antibody finding without the typical complex of symptoms were not included in the measles group.
All 162 patients with clinical measles and the 20 patients with a diagnostic antibody finding but without the typical clinical picture and without a history of recent vaccination were included for further analysis in the measles group, which thus consisted of a total of 182 patients (52% of the total of 350 ALRI patients studied).
The ages of the 182 patients with measles ranged from 4 to 59 months with a median of 17 months. The majority (63%) were <2 years old and 24 (13%) were younger than 9 months, the recommended age for measles vaccination. Three children were younger than 6 months old, all of them diagnosed clinically and one with a diagnostic antibody finding as well. The male to female ratio was 1.6:1. Seventy-five percent were malnourished: 47% wasted, 10% stunted; and 18% wasted and stunted. Eight and 4% had weight for height and height for age Z scores of −4 SD or less, respectively.
Rash was absent in 10 patients with a diagnostic antibody response. All except one were malnourished with >−2 SD Z scores for either weight for height or height for age and 1 was <−4 SD for weight for height. Four had been given measles vaccine. Seven of the rashless patients were diagnosed to have severe pneumonia whereas 3 had very severe pneumonia.
Of the 182 patients 166 (91%) had not received measles vaccine whereas the guardians of 12 (7%) reported that the child had been immunized against measles. Four guardians were uncertain about the immunization history. During the study period the monthly proportion of measles cases of all pneumonia cases enrolled varied from 30 to 81%. The peak occurrence was from March to June.
Pulmonary infiltrates were present radiographically in 151 measles patients (85%) including 36% with dense infiltrates (Table 2). Diarrhea, stomatitis and oral thrush were common.
Infection with bacteria or viruses other than measles was recognized in 97 (53%) of the 182 measles patients. Blood culture was positive in 10 patients (5.5%): S. pneumoniae in 5 patients, H. influenzae in 3 and Staphylococcus aureus in 2.
Another viral agent was implicated in 95 measles patients. In 60 patients 1 other viral infection was detected, and in 35 there were 2 or more. A total of 149 nonmeasles virus infections were identified (Table 3). The most common concomitant virus was parainfluenza virus (31% of virus infections), followed by adenovirus (23%). Almost one-half of the adenovirus infections were positive in both serology and culture or antigen detection. For the other respiratory virus infections, the diagnosis was based on serology in most cases. There were 8 patients with evidence of bacteria and at least 2 viruses (measles and another virus).
Thirty-one patients (17%) with measles-associated pneumonia died; however, the outcome of 7 patients was unknown because of incomplete follow-up in the community. Thirty-nine percent (12 of 31) of those who died were younger than 1 year old. Factors associated with a significantly increased odds ratio (OR) for dying from measles were cyanosis(OR 4.7, 95% CI 1.7 to 13), respiratory rate ≥60/min on admission (OR 3, 95% CI 1.3 to 7) and admission assessment fulfilling the criteria for very severe pneumonia(OR 5.3, 95% CI 2.3 to 12). Age younger than 1 year, presence of a dense infiltrate on chest radiograph, malnutrition and bacteremia were not associated with significantly increased OR.
There was a tendency toward a higher case fatality rate among patients with measles plus bacteria (3 of 10, 30%) than in those with measles only(21 of 80, 26%) and those with measles plus other viruses (7 of 85, 8%), but this difference was not significant. There was significantly more often a dense radiographic infiltrate among patients with measles and bacterial infection than in those with other viruses (P = 0.007, Fisher's exact test) and those with measles only (P = 0.004, Fisher's exact test) (Table 4).
Earlier hospital-based studies in the Philippines have shown that almost one-half of patients admitted for pneumonia have concomitant measles.11, 12, 26 In our prospective study on Filipino children younger than 5 years old with ALRI from a semiurban community where most families belong to the lower socioeconomic levels, we found that the disease still accounted for more than one-half of the pneumonia cases and 57% of all deaths among our hospitalized patients with ALRI. More than 10% of the cases were below 9 months of age, the age when measles vaccination is recommended.
To identify measles a sensitive antibody assay was used in addition to the clinical picture. It was positive in 96% of those with clinical measles and serum samples available. It also identified a significant number of measles patients(20 of a total of 182) in whom the complex of signs and symptoms typical for measles was absent.
The development of the rash in measles is a manifestation of functioning cell-mediated immunity.27 Among immunocompromised patients severe giant cell pneumonia caused by measles has been documented in the absence of a rash.28-31 The absence of the rash in a number of our cases might have been the result of the depressed immune response associated with malnutrition.
Measles-associated pneumonia can be caused by the measles virus itself, as well as by other viruses and by bacteria. Blood and lung puncture cultures done in measles patients have shown that pneumococcus, H. influenzae and Staphylococcus aureus14-16 are frequent concomitant causes of pneumonia. Adenovirus and herpes simplex virus are the most common viruses detected by lung puncture and in autopsy material13-16 in fatal cases of measles. Mixed infection was identified in more than one-half of our patients with measles associated pneumonia. Bacteria were found in 5% of blood cultures. The widespread practice in the Philippines of antibiotic use without prescription contributes to the low yield of blood cultures.32 We found concomitant viral infection in as many as 95 (52%) of the measles patients. Parainfluenza viruses were the most commonly detected agents followed by adenovirus.
Previous studies have reported an increased risk of mortality among malnourished patients with measles pneumonia,5, 7 particularly among the young.4, 33 The present study, however, did not support this finding.
Despite the introduction of single dose measles vaccination into the Philippine Expanded Programme of Immunization 7 years before the study, the proportion of pneumonia caused by measles remained high among our hospitalized population.
Financial support for this project was provided by the Academy of Finland and the Sigrid Juselius Foundation. We acknowledge the contribution of Ms. Terttu Autio for the measles antibody assay and Ms. Elinor Sunico and Ms. Ana Tamundong for the data analysis.
1. Garenne M, Ronsmans C, Campbell H. The magnitude of mortality from acute respiratory infections in children under 5 years in developing countries. World Health Stat Q 1992;45:180-91.
2. Markowitz LE, Nieburg P. The burden of acute respiratory infections due to measles in developing countries and the potential impact of measles vaccine. Rev Infect Dis 1991;13(Suppl 6):S555-61.
3. Aaby P, Bukh J, Lisse IM, Smits A. Measles mortality, state of nutrition and family structure: a community study from Guinea-Bissau. J Infect Dis 1983;147:693-700.
4. Aaby P, Bukh J, Lisse IM, et al. Determinants of measles mortality in a rural area of Guinea-Bissau: crowding, age and malnutrition. J Trop Pediatr 1984;30:164-8.
5. Walsh J. Selective primary health care: strategies for control of diseases in the developing world: IV. Measles. Rev Infect Dis 1983;5:330-40.
6. Tupasi TE, De Leon LE, Lupisan S, et al. Patterns of acute respiratory tract infections in children: a longitudinal study in a depressed community in Metro Manila. Rev Infect Dis 1990;12(Suppl II 8):S940-1054.
7. Guillozet N. Measles in Africa: a deadly disease. Clin Pediatr 1979;18:95-100.
8.Morley D. Severe measles in the tropics: I. Br Med J 1969;1:297-300.
9. Cherry JD. Measles. In: Feigin RD, Cherry JD, eds. Textbook of pediatric infectious diseases. 3rd ed. Philadelphia: 1992:1591-609.
10. Tidstrom B. Complications in measles with special reference to encephalitis. Acta Med Scand 1968;184:411-15.
11. Tupasi TE, Lucero MG, Magdangal DM, et al. Etiology of acute lower respiratory tract infections in children from Alabang, Metro Manila. Rev Infect Dis 1990;12(Suppl 8):S929-39.
12. Ruutu P, Halonen P, Meurman O, et al. Viral lower respiratory tract infections in Filipino children. J Infect Dis 1990;161:175-9.
13. Kipps A, Kaschula ROC. Virus pneumonia
following measles: a virological and histological study of autopsy material. S Afr Med J 1976;50:1083-8.
14. Morton R, Mee J. Measles pneumonia
: lung puncture findings in 56 cases related to chest x-ray changes and clinical features. Ann Trop Pediatr 1986;6:41-5.
15. Wesley AG, Sutton JB, Widrich AJ. The aetiology of pneumonia
associated with measles in Bantu children. S Afr Med J 1971;45:1402-4.
16. Dover AS, Escobar JA, Duenas AL, Leal EC. Pneumonia
associated with measles. JAMA 1975;234:612-14.
17. Acute respiratory infections in children: case management in small hospitals in developing countries. Programme for the control of ARI. Geneva: World Health Organization, 1988.
18. Acute respiratory infections. In: Laboratory manual of bacteriologic procedures. WHO Western Pacific Education in Action Series. Geneva: WHO, 1986.
19. Hierholzer JC, Johansson KH, Anderson LJ, Tsou CJ, Halonen PE. Comparison of monoclonal time-resolved fluoro-immunoassay with monoclonal capture-biotinylated detector enzyme immunoassay for adenovirus antigen detection. J Clin Microbiol 1987;25:1662-7.
20. Hierholzer JC, Bingham PG, Coombs RA, Johansson KH, Anderson LJ, Halonen PE. Comparison of monoclonal antibody time-resolved fluoro-immunoassay with monoclonal antibody capture biotinylated detector enzyme immunoassay for respiratory syncytial virus and parainfluenza virus antigen detection. J Clin Microbiol 1989;27:1243-9.
21. Scalia G, Halonen PE, Condorelli F, Mattila ML, Hierholzer JC. Comparison of monoclonal biotin-avidin enzyme immunoassay and monoclonal time-resolved fluoroimmunoassay in detection of respiratory virus antigens. Clin Diagn Virol 1995;3:351-9.
22. Koskinen P, Vuorinen T, Meurman O. Influenza A and B virus IgG and IgM serology by enzyme immunoassays. Epidemiol Infect 1987;99:55-64.
23. Vuorimaa T, Arstila P, Ziola B, Salmi A, Hanninen P, Halonen PE. Solid-phase radioimmunoassay determination of virus-specific IgM antibody levels in a follow-up of patients with naturally acquired measles infections. J Med Virol 1978;2:271-8.
24. Meurman O, Ruuskanen O, Sarkkinen H, Hanninen P, Halonen PE. Immunoglobulin class-specific antibody response in respiratory syncytial virus infection measured by enzyme immunoassay. J Med Virol 1984;14:67-72.
25. WHO Working Group. Use and interpretation of anthropometric indices of nutritional status. Bull WHO 1986;64:929-41.
26. Tupasi TE, Velmonte MA, Sanvictores MEG, et al. Determinants of morbidity and mortality due to acute respiratory infections: implications for intervention. J Infect Dis 1988;157:615-23.
27. Breitfeld V, Sherman FE, Odagiri K, Yunis EJ. Fatal measles infection in children with leukemia. Lab Invest 1973;28:279-91.
28. Enders JF, McCarthy K, Mitus A, Cheatham WJ. Isolation of measles virus at autopsy in giant cell pneumonia
without a rash. N Engl J Med 1959;261:875-81.
29. Markowitz LE, Chandler FW, Roldan EO, et al. Fatal measles pneumonia
without rash in a child with AIDs. J Infect Dis 1988;158:480-3.
30.Kaplan LJ, Daum RS, Smaron M, McCarthy CA. Severe measles in immunocompromised patients. JAMA 1992;267:1237-41.
31. Katz M, Stiehm ER. Host defense in malnutrition. Pediatrics 1977;59:490-4.
32.Lansang MA, Lucas-Aquino R, Tupasi TT, et al. Purchase of antibiotics without prescription in Manila, the Philippines: inappropriate choices and doses. J Clin Epidemiol 1990;43:61-7.
33. Scheifele DW, Forbes CE. The biology of measles in African children. East Afr Med J 1973;50:169-73.