A study of pneumonias in relation to bacterial flora and response to treatment in adultsGH Tilve, SR Kamat, VR Hoskote, Parvin S Anklesaria, GH Kashyap, UK Sheth
Department of Medicine, Chest Medicine and Clinical Pharmacology, Seth G.S Medical College and K.E.M. Hospital, Bombay 400 012, India
Correspondence Address: Source of Support: None, Conflict of Interest: None PMID: 572425
Source of Support: None, Conflict of Interest: None
A study of 93 cases of Pneumonias in adults was carried out over a period o f 3 years. The predominant organisms isolated were Enterobacter, E. coli, Streptococci and Staphylococci. Clinical and radiographic responses to the drugs administered (Epicillin, Ampicillin and Penicillin) were broadly comparable. Only 5 cases (Epicillin) failed to respond. There was little difference in the responses according to the organisms isolated or the sensitivity to the drug administered. The responses were similar in both Gram positive and Gram negative infections and the drug sensitivity status did not seem to matter.
Organisms isolated from sputa at our Institute , in recent years in cases with lobar or segmental pneumonias were different from those reported by other workers in India. , So, a prospective study of bacteriological isolations, clinical presentation and response to treatment was undertaken in January 1972 and continued till December 1974.
Over a three-year period, 108 cases with a history of fever, respiratory symptoms and radiographic evidence of segmental or lobar consolidation who were admitted to any medical unit of this large general hospital were studied. Out of these, a total of 15 cases (a) with a suspicion of malignancy, (b) obvious tuberculosis, (c) who had received antibiotics or chemotherapeutic drugs in previous 3 days, (d) who were pregnant, (e) who were younger than 12 years and (f) who were moribund, were excluded because of the difficulties in assessment [Table 1].
Each case was fully assessed clinically, biochemically and radiographically. The sputum collection was carried out by a trained doctor who ensured that the secretions were from the lower respiratory tract. The sputum was collected after the patient had cleaned his mouth, cleared nasal secretions and had gargled with sterile saline. After this, he spat directly in an autoclaved Petri dish More Details which was then transported to the laboratory within 10 minutes. The sputum smear was examined for the predominant organisms by Gram staining and the flora were categorised as Gram positive, Gram negative or mixed. The specimen was cultured by direct plating and drug sensitivity was carried out by the disc method on the pure culture isolated at 24 hours.  The radiograph was repeated after 7-10 days. At this stage the patient was discharged and followed up at 3-4 weeks for a further clinical and radiographich status. At the end of the trial, all improvements were graded jointly by two senior doctors.
The study was carried out on patients in 3 phases. The first series of 36 cases was done over 7 months in 1972 and were given oral Epicillin [6 (1) - 2 amino (1.4 cyclohexadiethyl) acetamido penicillanic acid) ] 500 mg. 4 times a day for a period of 7-10 days. The second series was done over 12 months in 1972-1973 in 21 cases and these were randomly given ampicillin or epicillin 500 mg. Q.I.D. orally as above. The third series of 36 cases were studied double blind over 9 months in 1974 to compare the efficacy of parenteral (IM) penicillin (1 million units 12 hourly) with that of epicillin 500 mg. 12 hourly.
In the first series, the sputum samples were collected again at 2-3 days after starting treatment in only 7 cases. But later this was regularly done with a view to check on the consistency of laboratory isolations.
[Table 1] gives the constitution of the study group, the number of cases excluded from the series and the reasons for exclusion.
There was a predominance of males (70 cases: 75%). Of 93 cases, 26 (28%) had an age of 40 years or more. Only 14 patients (15%) had a history of chronic chest illness in the past.
30 cases had a temperature of 39°C or higher on admission. In 54% of the cases the lesion on a radiograph was in the right lung, whereas both the lungs were involved in 8%.
[Table 2] lists the organisms that were isolated. The predominant organisms in the first series were Enterobacter, Streptococci and Colif orms. In the second series, they were Enterobacter and Klebsiella while in the third series Streptococci, Enterobacter, Staphylococci and Colif orms were commonly found. Pneumococci were isolated in only 4 subjects.
For checking on our methodology of sputum collection and processing, we did cultures on two occasions (at the intervals of 48 to 72 hours) in 20 normals, 20 patients of acute bronchitis (controls) and in 50 subjects with acute infective exacerbations and chronic lung disease. The predominant organisms isolated were Streptococcus pyogenes, Branhamella, catarrhalis and Staphylococcus. In 33 of these we isolated same organisms on duplicate testing. These aspects are being discussed separately in another communication. 
In 6 cases, we cultured material obtained by lung puncture aspiration. In only one case, the same organism (Staphylococcus albus) was isolated. In 2, no pathogens were isolated and in the remaining, Enterobacter aerogenes, Alkaligenes faecalis and Pseudomonas were isolated. As one of these patients died after 1 day, we did not pursue this method.
As shown in [Table 3], clinical response was assessed at 7-10 days (early) in 93 patients and after discharge from hospital at 3-4 weeks (late) in 55 patients. The improvement was graded as poor (no change), moderate (significant change in signs and symptoms) and good (complete disappearance of symptoms and signs.
For early response (upto 31%) in various series for either drugs, the differences were insignificant. The late response to epicillin was good in 29 (70%) cases. In comparison, 13 (93%) of 14 cases followed on the other two drugs showed good response (p > 0.05).
[Table 4] shows the correlation of the early clinical response with bacteriologic isolations. The isolates have been classified and grouped as (a) only Gram positive (Streptococcus, Staphylococcus and Pneumococcus), (b) only Gram negative subcategorised as: (i) E. coli and Enterobacter, (ii) other Gram -ve viz. Pseudo monas, A. anitratus, Proteus (all as single isolates) and (iii) Multiple organisms. When both Gram positive and negative organisms were cultured these were classified as (c) mixed.
In 19 patients with only Gram positive infections, 6 (32%) showed good clinical response. By contrast, of 47 cases with Gram negative infection, 13 showed good response. In 25 cases with mixed infection 6 (24%) showed good response.
As shown in [Table 5], the radiographic response was assessed as poor, slight, moderate and good. Thus good response at early stage was seen in 39 (59%) of 67 patients on epicillin, 7 (78 % ) of 9 patients on ampicillin and 8 (50%) of 17 patients on penicillin. The differences are significant (p < 0,05). The respective figures for late response were 36 (90%) epicillin, 5 (83%) ampicillin and 6 (86%) penicillin in 55 cases who were followed. When the early response in these 55 was studied it was found that there was a significant further improvement from 10 to 30 days in the radiographs (p < 0.05).
[Table 6] relates early radiographic response to bacteriologic isolations. Irrespective of the type of organisms cultured, there was good response in about 50 to 60% cases but there were no significant differences.
Correlation of the clinical and radiographic response is shown in [Table 7]. It is seen that a significant number (58%) with good radiographic improvement did not have good clinical response. But the overall response by two criteria was similar.
When the radiographic response was correlated with the degree of rise in temperature (above 39°C) on the first day or the age there were no differences. There were also no differences in the response in older subjects.
Besides 15 exclusions, only 2 (both on epicillin) were given another antibiotic. One was given tetracycline and another a combination of chloramphenicol with erythromycin. Both responded well. One case (age 45 years) on epicillin died on the 11th day with pyogenic meningitis and pneumonia (as confirmed by autopsy). One case, after fair radiographic response, died on the 18th day of an unrelated cause.
These were observed in 8 cases (5 epicillin, 2 ampicillin, 1 penicillin). One case on penicillin showed skin rashes, petechial haemorrhages and conjunctivitis. One case on ampicillin showed erythematous rash on the trunk on the 9th day, while another had itehing. Two cases on epicillin had vomiting and giddiness, while one case had diarrhoea; urticaria with itching was observed in one case and depigmentation of terminal phalanges with exfoliation was observed in one case. In none, the drug schedule was changed.
Our experience reveals the pattern of pneumonias in an Indian general hospital during the last 3 years. Of 108 cases, 8 were detected to have active tuberculosis presenting as lobar consolidation. Of 4 deaths two occurred in the first 24 hours, in two there was late death. In 2 others there was clinical deterioration; thus collective failure was seen in 6 per cent. Of the 5 uncooperative patients, none appeared resistant to treatment. None of the drugs studied gave differing responses. As compared to other Indian reports , who reported Pneumococci, Staphylococci and Klebsiella as dominant bacteria, our results revealed these to be Enterobacter, Streptococci and E. coli. While Pneumococci seemed uncommon, there were differences in isolations over 3 years. Barrett-Conner , has commented on the difficulty of isolating Pneumo cocci from sputum even in the presence of a systemic bacteremia. She quotes earlier reports stating that Pneumococci are distributed in a sputum specimen unevenly and may not be isolated consistently. While washing of the purulent portion of sputum is suggested for excluding the throat organisms, no systematic studies have been done to assess its effects. We do not approve of lung puncture or intratracheal aspiration as a routine procedure because of the inherent risks. As per criteria suggested by Rose et al  to declare an isolate pathogenic, one should obtain a confirmation in the culture from either another sample of sputum, blood, lung or pleural fluid. Of the 64 cases where duplicate cultures were done, only in 25 the same organisms were isolated. The organisms were Enterobacter (8), Klebsiella (2), Streptococcus (7), Strep tococcus with Staphylococcus (20), Stap hylococcus (2), Pneumococcus (1), Proteus (1), B. catarrhalis (2), no pathogen (1). We submit on this evidence that the predominant organisms causing pneumonias in India have now changed. Perhaps a plausible cause may be widespread use of antibiotics in urban areas. This may also explain why a significant proportion (43 per cent) revealed multiple organisms.
Comparable work has been reported from southern U.S.A. by Sullivan et al  in 1972. It reviews 292 cases of pneumonias in adults. Only in 167 cases could they attribute the causes to a particular pathogenic bacillus. The causative organisms were Pneumococcus in 62%, Staphylococcus in 10% and Gram negative bacteria in 20%. They studied viral and mycoplasmal cultures in all cases and found that in adult segmental pneumonias, virus infections appeared to be rare. A great majority had many associated chest diseases and overall mortality was 24 per cent. Those with age above 40 and with Gram negative bacilli had slightly poorer prognosis.
Sainani and Fulambarkar  isolated organisms consistent with sputum cultures from lung biopsy cultures, Of 6 lung aspirates of our series we could isolate the same organisms only once.
Bacterial isolations in control subjects were mainly Streptococci, Branhamella catarrhalis and Staphylococci. But those with acute lower respiratory infections showed organisms similar to those seen in pneumonia cases.
We thank Dr. C. K. Deshpande, Dean, for allowing us to carry out this work. Kind help in carrying out this work was given by Prof. K. G. Nair, Dr. A. B. Shah, Dr. P. J. Mehta. We wish to offer grateful thanks to Dr. H. C. Barbhaiya of Sarabhai Chemicals for the supply of the drugs and all the required help
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7]