66

tion in the management of acute myeloid leukemia in first remission. N Engl J Med1998;339: 1649-56.

27 Burnett AK, Goldstone AH, Stevens RMF, Hann IM, Rees JK, Gray RG, et al.Randomised comparison of addition of autologous bone- marrow transplantation tointensive chemotherapy for acute myeloid leukemia in first remission: Results ofMRC AML 10 trial. Lancet 1998;351:700--8.

28 Antman KH, Armitage JO, Horowitz MM, Rowlings PA. Autotransplants for breastcancer in North America. Proc Annu Meet Am Soc Clin OncoI1994;13:A69.

29 Cannellos GP. Selection bias in trials of transplantation for metastatic breast cancer:Have we picked the apple before it was ripe? J Clin OncoI1997;15:3169-70.

30 Rahman ZU, Frye DK, Buzdar AU, Smith TL, Asmar L, Champlin RE, et al. Impactof selection process on response rate and long-term survival of potential high-dosechemotherapy candidates treated with standard-dose doxorubicin containing chemo-therapy in patients with metastatic breast cancer. J c/in OncoI1997;15:3171-7.

31 Garcia-Carbonero R, Hindalgo M, Paz-Ares L, Calzas J, Gomez H, Guerra JA, etal. Patient selection in high-dose chemotherapy trials: Relevance in high-risk breastcancer. J Clin OncoI1997;15: 3178-84.

32 Gratwohl A, Hermans J, Baldomero H. Hematopoietic precursor cell transplants inEurope: Activity in 1994. Report from the European Group for Blood and MarrowTransplantation (EBMT). Bone Marrow Transplant 1996;17:137-48.

33 Kumar L, Gulati SC. Peripheral stem-cell transplantation.Lancet I 995;346 (Suppl):9.34 Roberts MM, To LB, Gillis D, Mindy J, Rawling C, Ngo K, et al. Immune

reconstitution following peripheral blood stem cell transplantation, autologousbone marrow transplantation and allogeneic bone marrow transplantation. BoneMarrow Transplant 1993;12:469-75.

35 Larsson K, Bjorkstrand B, Ljungman B. Faster engraftment but no reduction ininfectious complications after peripheral blood stem cell transplantation comparedto autologous bone marrow transplantation. Support Care Cancer 1998;6:378-83.

36 Nemunaitis J, Buckner CD, Dorsey KS, Willis D, Meyer W, Applebaum F. Retro-spective analysis of infectious disease in patients who received recombinant humangranulocyte-macrophage colony-stimulating factor versus patients not receiving acytokine who underwent autologous bone marrow transplantation for treatment oflymphoid cancer. Am J Clin OncoI1998;21:341--6.

37 Geisler CH, Hansen MM, Andersen NS, Broun P, Christensen LD, Dickmeiss E,etal. BEAM+autologous stem cell transplantation in malignant lymphoma: 100 con-secutive transplants in a single centre. Efficacy, toxicity and engraftment in relationto stem-cell source and previous treatment. Eur JHaematoI1998;61:173-82.

THE NATIONALMEDICAL JOURNAL OF INDIA VOL. 13, NO.2, 2000

38 Socinski MA, Cannistra SA, Elias A, Antman KH, Schnipper L, Griffin JD, et al.Granulocyte-macrophage colony stimulating factor expands the circulating haemo-poietic progenitor cell compartment in man. Lancet 1988;1:1194-8.

39 Sheridan WP, Begley CG, Juttner CA, Szer J, To LB, Maber D, et al. Effect ofperipheral blood progenitor cells mobilised by filgastrim (G-CSF) on plateletrecovery after high-dose chemotherapy. Lancet 1992;339:640-4.

40 Schwartzberg LS, Birch R, Hazelton B, Tauer KW, Lee P Jr, Altemose R, et al.Peripheral blood stem cell mobilization by chemotherapy with and without recombi-nant human granulocyte colony-stimulating factor. JHematother 1992;1:317-27.

41 Brandt SJ, Peters WP, Atwater SK, Kurtzberger J, Borowitz MJ, Jones RB, et al.Effect of recombinant human granulocyte-macrophage colony stimulating factoron hematopoietic reconstitution after high-dose chemotherapy and autologous bonemarrow transplantation. N Engl JMed 1988;318:869-76.

42 Sheridan WP, Morstyn G, Wolf M, Dodds A, Lusk J, Maher D, et al. GranulocyteCOlony-stimulating factor and neutrophil recovery after high-dose chemotherapyand autologous bone marrow transplantation. Lancet 1989;2:891-5.

43 Nemunaitis J, Rabinowe SN, Singer JW, Bierman PJ, Vose JM, Freedman AS, eta I.Recombinant granulocyte-macrophage colony-stimulating factor after autologousbone marrow transplantation for lymphoid cancer trials. N Engl JMed 1991;324:1773-8.

44 Stiff PJ, Dahlberg S, Forman SJ, McCall AR, Horning SJ, Nademanee AP, et al.Autologous bone marrow transplantation for patients with relapsed or refractorydiffuse aggressive non-Hodgkin's lymphoma: Value of augmented preparativeregimens-a Southwest Oncology Group Trial. J c/in OncoI1998;16:48-55.

45 Moreau P, Milpied N, Rapp MJ, Morseau A, Bourdin S, Mahe MA, et al. Earlyintensive therapy with autologous stem cell transplantation in high-risk Hodgkin'sdisease: Long term follow up in 35 cases. Leuk Lymphoma 1998;30:313-24.

46 Moreau P, Fleury J, Brice P, Colombat P, Bouabdallah R, Lioure B, et al. Earlyintensive therapy with autologous stem cell transplantation in advanced Hodgkin'sdisease: Retrospective analysis of 158 cases from the French registry. Bone MarrowTransplant 1998;21:787-93.

47 Arranz R, Tomas JF, Gil-Fernandez JJ, Martinez Chamorro C, Granados E, et al.Autologous stem cell transplantation (ASCT) for poor prognosis Hodgkin's disease(HD): Comparative results with two CBV regimens and importance of disease statusat transplant. Bone Marrow Transplant 1998;21:779-86.

48 Andre M, Henry-Amar M, Blaise D, Colombat P, Fleury J, Milpied N, etal. Treat-ment -related deaths and second cancer risk after autologous stem-cell transplantationfor Hodgkin's disease. Blood 1998;92: 1933-40.

Mycoplasma pneumoniae and community-acquired.pneumoma

A. B. DEY, RAM A CHAUDHRY, P. KUMAR, NAZIMA NISAR, KALPANA M. NAGARKAR

ABSTRACTBackground. Community-acquired pneumonia is an impor-

tant cause of mortality and hospitalization in all age groups. Intemperate climates, Mycoplasma pneumoniae is a common respi-ratory pathogen causing pneumonia. Information on humanMycoplasma infection in India is scarce.

Hethods. We aimed to determine the frequency of Myco-

All India Institute of Medical Sciences, New Delhi 110029, IndiaA. B. DEY, P. KUMAR, KALPANA M. NAGARKAR

Department of MedicineRAMA CHAUDHRY, NAZIMA NISAR

Department of Microbiology

Correspondence to A. B. DEY

© The National Medical Journal of India 2000

plasma pneumoniae infection among patients with community-acquired pneumonia in a prospective cross-sectional study. Theassessment included clinical and radiological evaluation followedby microbiological evaluation for the specific pathogen. Micro-biological investigations included aerobic and anaerobic bloodculture, anti-Mycoplasma IgM antibody detection by gelatinparticle agglutination test and ELISA, culture of respiratory tractsecretions for Mycoplasma pneumoniae and other organisms, anddetection of specific Mycoplasma pneumoniae antigen by indirectimmunofluorescence.

Results. Sixty-two patients (42 men and 20 women; meanage 41.7 years) with community-acquired pneumonia wereinvestigated prospectively. They included 42 immunocompetentand 20 immunocompromised patients. Sixpatients had definitive

DEY et al. : Mycoplasma pneumoniae AND COMMUNITY-ACQUIRED PNEUMONIA 67

evidence of Mycoplasma pneumoniae infection and an additional16 patients had a probable diagnosis. In all, 22 (35.5%) patientswith pneumonia hadMycoplasma pneumoniae infection. Of these,12 patients belonged to the immunocompromised group and 10to the immunocompetent group. Patients with Mycoplasmapneumoniae infection also had secondary bacterial infection asevidenced by organisms isolated from blood in 50% and fromrespiratory tract secretions in 68%.

Conclusion. Community-acquired pneumonia has a poly-microbial aetiology, of which the prevalence of Mycoplasmapneumoniae is 35%. The study has two implications: (i) Myco-plasma pneumoniae infection is frequently associated with second-ary bacterial infection; and (ii) initial empirical antibiotic therapyfor community-acquired pneumonia in India must include anti-biotics with activity against Mycoplasma pneumoniae.

Natl Med J India 2000; 13 :66-70

INTRODUCTIONCommunity-acquired pneumonia or lower respiratory tract infec-tion is a common and important cause of hospitalization and deathin all age groups. As several organisms have been identified ascausative agents, choosing an antibiotic for the treatment oflowerrespiratory tract infection is often challenging. The task hasbecome even more daunting as a result of the changing spectrumoflower respiratory tract infection and development of drug resis-tance among the common pathogens.

Mycoplasma pneumoniae (M. pneumoniae) is a common res-piratory pathogen in temperate climates. However, its epidemiol-ogy is not clearly defined due to difficulty in establishing a defini-tive diagnosis. 1Laboratory diagnosis of Mycoplasma infectionrequires isolation of the organism by culture. This may take up to8 weeks and has a sensitivity of 57%.2.3 Detection of anti-Mycoplasma IgM antibody by the complement fixation test(Cf'T)" or cold agglutination' were previously the commonly usedmethods for diagnosis of acute infection. In recent years, sensitivetechniques such as ELISA6 and indirect immunofluorescence?have been adopted for the detection of anti-Mycoplasma IgMantibodies in blood andM. pneumoniae specific antigen in respi-ratory tract secretions.s?

Information on human Mycoplasma infection in a tropicalclimate such as that ofIndia is scanty. In the 1970s, using serologi-cal methods for diagnosis, the prevalence of M. pneumoniaeinfection was found to be 14%10in respiratory tract infections and16%11in atypical pneumonia. In a recent report involving 400patients, anti-M. pneumoniae IgM antibodies were present in 26%of sera samples whereas culture positivity was reported in 10% ofrespiratory tract secretion specimens."

In the present study, the magnitude of M. pneumoniae infec-tion was determined by isolation and antigen and antibodydetection among immunocompetent and immunocompromisedpatients with community-acquired pneumonia.

PATIENTS AND METHODSPatientsIn a prospective study, carried out over a period of 9 months,consecutive immunocompetent and immunocompromised pa-tients (aged 12years or more) admitted with community-acquiredpneumonia to a medicine unit ofthe All India Institute of MedicalSciences, New Delhi were investigated. Patients were consideredimmunocompromised if they were receiving corticosteroids inhigh doses, immunosuppressive drugs after renal transplant,

chemotherapy for haematological and non-haematological ma-lignancies, or were suffering from end-stage chronic renal failure.Patients were considered immunocompetent when the above-mentioned conditions were absent along with all other knowncauses of immunosuppression, including human immunodefi-ciency virus (HIV) infection. The criteria for diagnosis of commu-nity-acquired pneumonia were adopted from Fang et al. 13

1. Presence of at least one major clinical criterion (cough, sputumproduction or fever >37.8 0C) or two minor criteria (pleuriticchest pain, dyspnoea, altered mental state, signs of pulmonaryconsolidation on examination or total leucocyte count> 12 000per cmm).

2. Presence of a new pulmonary infiltrate/shadow on chest X-raysuggestive of pneumonia at orwithin 24hours of hospitalization.

3. Patient residing in the community.

The criteria for exclusion from this study were: 13

1. Pulmonary shadow due to a cause other than pneumonia, and2. Patient brought from another hospital or with a history of

hospitalization within 7 days of discharge.

ControlsTwenty-three adults were included in the study as controls forserological comparison and were recruited from the medical out-patient department. Controls were patients presenting with non-specific and/or psychosomatic complaints. They had no evidenceof pneumonia, or any infectious or respiratory disease.

EvaluationA history was taken and physical examination carried out in allpatients to establish the clinical diagnosis of pneumonia. Labora-tory investigations done in these patients included:

1. Chest X-ray (PA view) to establish the diagnosis of pneumonia.2. Complete blood count and routine blood chemistry to evaluate

the clinical status.3. Aerobic and anaerobic blood culture.4. Aerobic sputum culture (after ensuring that the specimen was

from the lower respiratory tract by screening with Gram-stain)5. Investigations to make a diagnosis of M. pneumoniae.

During bacteriological evaluation of body fluids, organismswere identified as described by Cruickshank." Aerobic organ-isms were isolated by inoculating clinical samples on sheep bloodagar and MacConkey agar plates. The plates were incubated at37°C aerobically under 5% CO2 for microphilic organisms; andwere examined at 24 and 48 hours. For anaerobic culture, thespecimens were plated into brain-heart infusion agar and en-riched thioglycolate broth. These media were incubated in anaero-bic jars and examined at 48 and 96 hours. Subcultures fromthioglycolate broth were inoculated on brain-heart infusion agarand incubated in an anaerobic jar. Biochemical identification andantibiotic susceptibility of the aerobic and anaerobic strains wereperformed by methods described previously.r':" All organismsisolated from blood were considered significant after exclusion ofcontaminants. Coagulase-negative Staphylococcus isolated inblood was further characterized. All strains of Staphylococcusepidermidis isolated from blood and sputum were consideredpathogenic in immunocompromised patients, especially whenisolated from both blood and sputum.

Microbiological investigations for diagnosis of M. pneumoniaeIsolation of the organism. Respiratory tract secretions (throat

68

swab) were collected, transported in PPLO broth and cultured onconventional Hayflick or SP4 medium. M. pneumoniae wasidentified on SP4 broth by glucose fermentation and furtherconfirmed on agar plates by typical colony appearance, Dienesstaining and inhibition of growth by specific anti-sera to M.pneumoniae.

Detection of M. pneumonia antigen? Respiratory secretions(throat swab) were spotted onto a glass slide, air-dried at roomtemperature and fixed; and then incubated with anti-M.pneumoniaeantibodies (rabbit polyclonal sera obtained from the CentralPublic Health Laboratory, London). Fluorescent conjugated anti-rabbit IgG antibody was added to the preparation which wasviewed for specific apple green fluorescence.

Detection of anti-M. pneumoniae antibody. Anti-M.pneumoniae IgM antibody was detected by the micro titre particleagglutination method (Serodia Myco II Reagent Kit, Fujirebio,Japan) which is based on the specific immuno-agglutinationprinciple. Presence of anti-M. pneumoniae antibodies in theserum led to agglutination of the gelatin particles visible with thenaked eye. The test was considered positive at a titre of 1:40 ormore. The positive results were reconfirmed by ELISA (Viro-ELISA, Viro Immun Labor Diagnostika GmbH). Anti-M.pneumoniae IgG antibodies were detected by ELISA (Viro-ELISA, Viro Immun Labor Diagnostika GmbH).

Diagnosis of M. pneumoniae infectionThe following criteria were used to classify the possibility of M.pneumoniae infection in the lower respiratory tract.

Definitive diagnosis1. Isolation of M. pneumoniae from cultures of respiratory

secretion, and/or2. Detection of M. pneumoniae antigen in respiratory secretions

with raised IgM (> 1:40) in the serum.

Probable diagnosis1. Detection of M. pneumoniae antigen in respiratory secretion

alone, or2. Raised IgM titre in serum alone3. Raised IgM and IgG titre in serum.

Past infection1. Detection of raised anti-M. pneumoniae IgG antibody titre in

serum2. Detection of raised anti-M. pneumoniae IgG antibody titre in

serum along with M. pneumoniae antigen in respiratory tractsecretions.

Statistical methodsWe used mean, standard deviation and median for descriptivestatistics and Chi-square test for comparative statistics.

RESULTSDemography and clinical featuresForty-two immunocompetent and 20 immunocompromisedpatients who fulfilled the defined criteria for community-acquiredpneumonia were investigated for the aetiology of pneumonia.There were 42 men and 20 women with a mean age of 41.7 years(range: 14 to 67 years) and included 11 teenagers (12 to 20 years).The mean duration of symptoms was 6.3 days (range 2-13 days;median 7 days) at the time of hospitalization.

After investigations, 22 (35.5%) patients were diagnosed to

THENATIONALMEDICALJOURNALOF INDIA VOL. 13, NO.2, 2000

have acute infection with M. pneumoniae. The symptoms andsigns of pneumonia in patients with and without M. pneumoniaeinfection are shown in Table I. There was no statistically signifi-cant difference in age and sex distribution, duration of illness,symptoms, signs and radiological features among the patients inthe two groups. M. pneumoniae infection was present in 4 of the11 teenagers. There was no mortality and all of them responded toempirical antibiotic therapy.

Microbiological diagnosis of Mycoplasma pneumoniae1. M. pneumoniae was isolated in 3 out of 30 patients in whom

respiratory secretions were cultured in SP4 medium.2. M. pneumoniae antigen was detected by indirect immuno-

fluorescence in 22 patients.3. IgM antibodies against M. pneumoniae were detected in 6

patients.4. IgG antibodies against M. pneumoniae were detected in 5

patients.

Using the diagnostic criteria defined earlier, 6 patients had adefinite diagnosis (3 in each group) and 16 patients (7 immuno-competent and 9 immunocompromised) had a probable diagnosisof acute M. pneumoniae infection. In addition, 4 patients hadevidence of past infection. Thus, 22 patients (35.5%) with com-munity-acquired pneumonia had M. pneumoniae infection. Asummary of the above investigations is presented in Table II.Other organisms isolated from blood and respiratory tract secre-tions are listed in Table III. Other organisms were isolated fromthe blood of patients (50%) who were diagnosed to have M.pneumoniae infection. In addition, 15 M. pneumoniae-positivepatients (68%) had other organisms in their respiratory tractsecretions. Of the 40M. pneumoniae-negative patients, 14 (35%)were blood culture positive and 27 (68%) had various bacteriaisolated from respiratory tract secretions. No anaerobic organismswere isolated from the blood or respiratory tract secretions.

ControlsAll the control sera were negative for IgM by the gelatin particleagglutination method with a cut-off titre of 1:40. The mean (SD)

TABLEI. Clinical features of patients with community-acquiredpneumonia

Characteristic M. pneumoniae

Absent(n=40)

Present(n=22)

Meanage (in years)CoughFeverExpectorationShortnessof breathChestpainHaemoptysisHypotensionChestsignsChestX-rayabnormality

Air spacepneumoniaLobar consolidationMulti-lobarconsolidationBronchopneumoniaInterstitialpneumoniaCavitarypneumoniaPleuraleffusion

41.5

3527271573539402921882oI

41.95202116962220221165542o

DEY et al. : Mycoplasma pneumoniae AND COMMUNITY-ACQUIRED PNEUMONIA 69

TABLEII. Investigation profile of M. pneumoniae-positive patients

Investigation Immunologically Total

(n=62)competent(n=40)

compromised(n=22)

Only culture 1 I 2Only antigen 7 7 15Only antibody (IgM) 0 1 1Only antibody (IgG) 0 0 0Antigen and antibody IgM 2 I 3Antigen and antibody IgG 3 I 3Antibody IgM and antibody IgG 0 I 1Culture, antigen and antibody IgM 0 I 1Acute infection 10 12 22Past infection 3 I 4

optical density of control sera for IgG antibodies was 0.171(0.005), which was less than the cut-off values advised by themanufacturer.

DISCUSSIONIn an analysis of 1445 patients included in 7 studies from temper-ate climates, M. pneumoniae was reported as a cause of commu-nity-acquired pneumonia with a relative frequency of 0% to 18%(mean 9%).1In an Indian study on lower respiratory tract infectionin children, M. pneumoniae infection was diagnosed in 27.4% ofcases by serological methods." In the present study, with thestated diagnostic criteria, 22 of 60 patients (35.5%) were consid-ered to have M. pneumoniae infection. The prevalence rate inimmunocompetent patients was 24% and in immunocompromisedpatients 55%. M. pneumoniae is generally believed not to causea very severe form of disease requiring hospitalization, especiallyin immunocompetent patients. Thus, it is possible that thesefigures are an under-estimation of the actual prevalence of thedisease as we did not study those who were treated as outpatients.

M. pneumoniae infection has been considered to occur moreoften in children and adolescents. However, in recent years, theorganism has been found to be a causative agent in all age groups.It has been reported as the causative agent of lower respiratorytract infection in up to 5% of older patients (mean age 65 years)"and the incidence increases with age, especially among hospital-ized patients. IS Thus, M. pneumoniae not only affects olderpatients but also causes more severe disease in them. In our studyalso, M. pneumoniae infection was encountered in all age groupsand the incidence was not higher in teenagers.

As for all microbes, the definitive diagnosis ofM. pneumoniaeinfection requires isolation of the organism, which is technicallydifficult, time intensive and has a low sensitivity." Most studieshave depended on the detection of rising titres of IgM antibodiesagainst the organism as the basis for serological diagnosis. Thevalue of serology is often limited in acute infection as develop-ment of specific anti-Mycoplasma IgM antibody requires theinfection to be present for several days. The use of paired serumspecimens collected at an interval of 2-3 weeks is ideal for theserodiagnosis of M. pneumoniae. However, there have beenseveral studies carried out using a single serum specimen for thediagnosis of M. pneumoniae. Hirai et al. considered a singleserum specimen CFT titre> 1:64 and GPA (gelatin particle agglu-tination) titre ~1 :320 as diagnostic of M. pneumoniae infection.'?In another study, single IgM EIA (enzyme immunoassay) ~1:60and IgG ~ 1:100 were considered diagnostic of current or recentinfection." We have also considered the presence ofIgM and IgG

TABLEIII. Organisms isolated from blood and respiratory tractsecretions in 62 patients with community-acquired pneumonia

Organism M. pneumoniae

positive (n=22) negative (n=40)*

BloodSterileContaminatedKlebsiella pneumoniaeStreptococcus pneumoniaeStreptococcusStaphylococcus epidermidisStaphylococcus albusCitrobacterAcinetobacterStaphylococcus aureusPseudomonasRespiratory tract secretionNormal floraKlebsiella pneumoniaeStreptococcus pneumoniaeStaphylococcus aureusHaemophilus influenzaeStaphylococcus epidermidisAcinetobacterPseudomonasCitrobacter

834

2242I222II2I

I4

714534

133351

251

* Two patients each had more than one organism isolated from blood and respiratorysecretions

antibodies as diagnostic of current infection. Warriset al. consid-ered an anti-M. pneumoniae IgG antibody in a single serumspecimen at a value21:40 U to be diagnostic by platelet EIA test."We considered the presence of IgG antibodies alone or with M.pneumoniae antigens as evidence of past infection. It has beenreported that following an acute infection, there is prolongedshedding of M. pneumoniae antigen from respiratory tract IgGantibodies in the serum. However, this is not true for IgM anti-bodies." In other words, detection of IgM antibodies is thecornerstone of serodiagnosis of currentM. pneumoniae infection.

IgM-specific EIA procedures have reduced the need for pairedspecimens for a more accurate diagnosis. We also used similartechniques to obviate the need for paired sera at the end of twoweeks. We used the GPA kit, which according to the manufacturerexclusively detects IgM antibody against M. pneumoniae and atitre of 1:40 is diagnostic of recent infection. This kit has beenpreviously compared with antibody-capture enzyme immuno-assay, which has been shown to be very sensitive and specific fordiagnosis."

Detection of antigens specific to M. pneumoniae using na-sopharyngeal aspirate and sputum has been found useful in thediagnosis and has a high sensitivity (91%).19Williamson et al.compared the sensitivity of antigen capture assay with detectionof M. pneumoniae DNA by PCR amplification-' and found that inup to 7.5% of samples, antigen capture assay was positive thoughPCR was negative. They postulated that the bacterial nucleic acidmight be degraded by nucleases in respiratory tract secretionsfollowing a breach in the cell membrane as a result of antibiotictherapy and development of immune defenses against the organ-ism.

The therapeutic utility of antigen detection would be veryhigh, especially in the early stages of infection when antibodyresponse is not detectable. We detectedM. pneumoniae antigen in19 patients with acute infection, in only 3 of whom anti-Myco-

70

plasma IgM antibody was present. The absence of specific IgMantibody in a substantial number of patients was probably due tothe investigation being done before the formation of such antibod-ies, as the median duration of illness in our patients at admissionwas only 7 days. Alternatively, in case of re-infection with M.pneumoniae an IgM antibody response may not be mounted at allby many adults. To the best of our knowledge, detection of M.pneumoniae antigen by indirect immunofluorescence has notbeen reported from India earlier.

Our approach of classifying M. pneumoniae infection as adefinite or probable cause of pneumonia has not been usedfrequently. A similar approach of classification has been used intwo epidemiological studies on pneumonia.P:" This approach canbe justified as most studies on community-acquired pneumoniafail to reach an aetiological diagnosis in as many as 50% ofpatients, despite excellent laboratory support. Pathogens otherthan M. pneumoniae were isolated from blood in 25 out of 62patients tested. The rate of isolation was 50% in patients with M.pneumoniae compared to 35% in M. pneumoniae-negativepatients. Similarly, in 68% of patients in both groups, otherpathogenic organisms were isolated from respiratory tract secre-tions.

In the present study, the clinical and radiological features ofpneumonia due to M. pneumoniae were indistinguishable fromthose due to other organisms. None of the patients had features of'atypical pneumonia'. It is likely that the inflammatory responsesof the respiratory tract to various infections are so limited thattheir clinical and radiological manifestations may not always bedistinctive. Alternatively, the clinical manifestations of M.pneumoniae infection were modified by the presence of otherrespiratory pathogens.

Polymicrobial aetiology has been reported as a common phe-nomenon in lower respiratory tract infections." It is possible, aswe also found, that in many patients with community-acquiredM.pneumoniae and pneumonia due to other atypical organisms, theprimary infection is followed by a secondary bacterial infection.The scenario is similar to viral lower respiratory tract infectionwith influenza virus or respiratory syncitial virus where second-ary bacterial colonization is a common occurrence. This results inthe loss of clinical and radiological distinctiveness of pneumoniadue to various organisms.

A limitation of the present study was that culture of M.pneumoniae could not be carried out in all patients and paired serawere not used. This might have resulted in higher prevalencerates. Despite this limitation, we have demonstrated the presenceof M. pneumoniae as a common cause of community-acquiredpneumonia especially in immunocompromised patients. Thisdiagnosis cannot be suspected on clinical and radiological grounds.

The therapeutic implication of our study is that antibiotics withanti-Mycoplasma pneumoniae activity must be included in allinitial empirical antibiotic regimens for community-acquiredpneumonia; that is, a combination of a beta-lactarn antibiotic witha second-generation macrolide, or a third-generation quinolone asa single agent. Further, in immunocompromised patients withpulmonary infection, this organism must be considered beforeinvasive procedures are undertaken to obtain an aetiologicaldiagnosis.

ACKNOWLEDGEMENTSThis study was partially supported by the Indian Council of Medical

THE NATIONAL MEDICAL JOURNAL OF INDIA VOL. 13, NO.2, 2000

Research grant no. 5/3/3/3/94 ECD-I. We gratefully acknowledge thetechnical help of Shri Sunil Kumar and Shri Salekh Chand.

REFERENCESKayser FH. Changes in the spectrum of organisms causing respiratory tractinfection: A review. Postgrad Med J 1992;68 (S3):S 17-S23.

2 Tully JG, Rose DL, Whitcomb RF, Wenzal RP. Enhanced isolation of Mycoplasmapneumoniae from throat washings with a newly modified culture medium. J InfectDis 1979;139:478-82.Kenny GE, KaiserGG, Cooney MK, Foy HM. Diagnosis of Mycoplasma pneumoniae:Sensitivities and specificities of serology with lipid antigen and isolation of theorganism on soy peptone medium for identification of infections. J Clin Microbiol1990;28:2087-93.

4 Kenny GE, Grayston JT. Eaton pleuro-pneumonia like organism (M. pneumoniaeycomplement fixing antigen: Extraction with organic solvents. J ImmunoI1965;95:19-25.

5 Denny FW, Clyde WA, Glezen WP. M. pneumoniae disease: Clinical spectrum,pathophysiology, epidemiology, and control. J Infect Dis 1971;123:74-92.

6 Uldum SA, Jensen JS, Sondergard-Andersen J, Lind K. Enzyme immunoassay fordetection of immunoglobulin M (lgM) and IgG antibodies to Mycoplasma pneu-moniae. J CUn MicrobioI1992;30:1198-204.

7 BarkerCE, Sillis M, WreghittTG. Evaluation of Serodia Myco II particle agglutinationtest for detecting M. pneumoniae antibody: Comparison with mu-capture ELISAand indirect immunofluorescence. J Clin Pathol 1990;43: 163-5.Kok TW, Varkanis G, Marmion BP, Martin J, Esterman A. Laboratory diagnosis ofM. pneumoniae infection. I. Direct detection of antigen in respiratory exudates byenzyme immunoassay. Epidemiol Infect 1988;101:669-84.

9 Harris R, Marmion BP, Varkanis G, Kok TW, Lunn B, Martin J, et al. Laboratorydiagnosis of M. pneumoniae infection. 2. Comparison of methods for the directdetection of specific antigen or nucleic acid sequences in respiratory exudates.Epidemiol Infect 1988;101:685-94.

10 Govil MK, Agrawal BD, Gupta BK, Pradhan S. M. pneumoniae and respiratorydiseases. J Indian Med Assoc 1978;70:269-72.

II Ayyagari A, Gupta U, Mohapatra LN, Balaya S, Bajaj S. A study of the role ofbacteria, viruses andM. pneumoniae in infants and children suffering from pneumonia.Indian J Med Res 1972;60: 1750-4.

12 Ramamoorthi U, Anand Rao U, Thyagrajan SP, Somu N, Balachandran A, AhmedB, et al. Mycoplasma pneumoniae in lower respiratory tract infection. Indian J MedMicrobioI1996;14:209-12.

13 Fang GD, Fine M, Orloff J, Arisumi D, Yu VL, Kapoor W, et al. New and emergingetiologies for community-acquired pneumonia with implications for therapy: Aprospective multi-centre study of 359 cases. Medicine (Baltimore) 1990;69:307-16.

14 Cruickshank R, Duguid JP, Marmion BP, Swain RHA. Medical microbiology.Edinburgh: ELBS and Churchill Livingstone, 1980.

15 Suiter VL, Citron OM, Edelstein MAC, Finegold SM (eds). wadswortn 's anaerobicbacteriology manual. St Louis:CV Mosby, 1985:23-67.

16 Chaudhry R, Nisar N, Dhawan B, Kabra SK. Prevalence of Mycoplasma pneumoniaeand Chlamydia pneumoniae in Indian children with community-acquired pneumonia.Indian Paediatr 1998;65:717-21.

17 Marston BJ, Plouffe JF, File TM Jr, Hackman BA, Salstrom SJ, Lipman HB, et al.Incidence of community-acquired pneumonia requiring hospitalization: Results ofa population-based active surveillance study in Ohio. The Community-BasedPneumonia Incidence Study Group. Arch Intern Med 1997;157:1709-18.

18 Ishida T, Hashimoto T, Arita M, Ito I, Osawa M. Etiology of community-acquiredpneumonia in hospitalized patients: A 3-year prospective study in Japan. Chest1998;114: 1588-93.

19 Hirai Y, Shiode J, Masayoshi T, Kanemasa Y. Application of an indirectimmunofluorescence test for detection of Mycoplasma pneumoniae in respiratoryexudates. J Clin MicrobioI1991;29:2007-12.

20 Warris ME, Toikka P, Saarinen T, Nikkari S, Meurman 0, Vainionpaa R, et al.Diagnosis of Mycoplasma pneumoniae pneumonia in children. J Clin Microbiol1998;36:3155-9.

21 Biberfield G. Antibody responses in Mycoplasma pneumoniae infection in relationto serum immunoglobulins, especially IgM. Acta Pathol Microbiol Scand B

Microbiollmmunol 1971 ;79:620-34.22 Echevarria JM, Leon P, Balfagon P, Lopez JA, Fernandez MV. Diagnosis of

Mycoplasma pneumoniae infection by microparticle agglutination and antibody-capture enzyme-immunoassay. Eur J Clin Microbial Infect Dis 1990;9:217-20.

23 Williamson J, Marmion BP, Worswick DA, Kok TW, Tannock G, Herd R, HarrisRJ. Laboratory diagnosis ofMycoplasmapneumoniae infection. 4. Antigen captureand PCR-gene amplification for detection of Mycoplasma: Problem of clinicalcorrelation. Epidemiol Infect. 1992;109:519-37.

24 Lieberman 0, Schlaeffer, Boldur I, Lieberman D, Horowitz S, Friedman MG, et al.Multiple pathogens in adult patients admitted with community-acquired pneumonia:A one-year prospective study of346 consecutive patients. Thorax 1996;51: 179-84.