key: cord-0007038-toycv4hg
authors: Mårdh, P. -A.; Hovelius, B.; Nordenfelt, E.; Rosenberg, R.; Soltesz, L. V.
title: The incidence and aetiology of respiratory tract infections in general practice — with emphasis onMycoplasma pneumoniae
date: 1976
journal: Infection
DOI: 10.1007/bf01638422
sha: 2117a7e64d0f127d9b6a652ef6df5852dc01974d
doc_id: 7038
cord_uid: toycv4hg

The incidence of respiratory tract infections in patients seeking medical advice at a community care centre (Dalby) during 1973 and 1974 was studied. About every third patient seen at this primary health station presented with signs of such infections. In the age groups <10, 10–19, 20–39, 40–59 and ≥60 years, respiratory tract infections accounted for 65, 45, 32, 18 and 9% of the total number of diagnoses made during 1974. The aetiology of acute respiratory tract infections in a series of patients seen at this health station was studied. The series included randomly selected cases, but excluded children under seven years of age and patients presenting with signs of acute otitis media and tonsillitis. Attempts to establish the aetiology were made on the basis of the history, the clinical examination, and cultures for beta-haemolytic streptococci andMycoplasma pneumoniae, complement fixation tests for influenza A and B, para-influenza 1, 2, and 3, adeno, cytomegalovirus and respiratory syncytial virus, andChlamydia psittaci. Paul-Bunnell test and tests for cold agglutinins were also performed. With this test battery, an aetiological diagnosis was obtained in only 33% of the 101 patients studied. The findings suggest an infection withM. pneumoniae in 16%, with beta-haemolytic streptococci in 9%, and with viruses (adeno and para-influenza) in 7% of the patients. The present communication highlights the role ofM. pneumoniae in upper respiratory infections, as few data have appeared on such infections in patients seen in general practice. The difficulty of establishing the aetiology of respiratory tract infections and the consequent treatment dilemma is discussed.

Symptoms of respiratory tract infection are one of the most common reasons for a patient to seek a consultation with his physician. The causative agents of respiratory tract infections are known to vary with the season, climate, age of the population, and with the epidemiological environment. Therefore the relative importance of the various agents involved varies widely. It is well known that viruses are the cause of the majority of these infections. More than 150 different viruses have so far been shown to cause infections ol the respiratory tract (17) . There are only minor differences between the clinical signs and symptoms of the infections caused by various viruses. This is also true of respiratory tract infections caused by bacteria, including those caused by beta haemolytic streptococci (4, 5) . It is therefore difficult to establish the aetiology on the basis of the clinical picture in any of these infections.

Mycoplasma pneumoniae is known as an important aetiological agent in pneumonia (11) but its relative importance in upper respiratory tract infection is still uncertain. In non-epidemic periods infections with this organism are known to be endemic. Epidemics seem to occur every fourth to fifth year (22) . Infections are seen in all ages but are more common in older children. The incubation period is usually three weeks or more. The contagiousness is relatively low and spread of the infection seems to occur after prolonged and close contact, for example within families (12) . A number of extra-respiratory tract manifestations have been described, including manifestations in the central nervous system (2, 28) , the skin and the mucous membranes, and in parenchymatous organs such as the liver and the pancreas (26) . It is rare for infections with M. pneurnoniae to be fatal.

The present study reports the results of a survey of the incidence and aetiology of respiratory tract infections in patients who were seen at a Community Care Centre. In this 

One of the objectives o£ the Community Care Sciences Centre at Dalby, Sweden, is to perform research in general practice. It is a primary health station serving an area with a mixed rural and urban population of about 15,000 inhabitants. These inhabitants visit the Centre about 10,000 times each year. About 40 % of the population is engaged in agriculture, and the remainder work at manufacturing, construction, in trade, or are in the public service. The village of Dalby is situated about ten kilometers from Lund, which is a university town with about 70,000 inhabitants. Dalby lies 24 kilometers from Maim6 with its 265,000 inhabitants. Many of those living in the Dalby district are employed in these two towns ( Figure 1 ). 

Clinical material One hundred and thirteen patients with signs and symptoms of acute respiratory infection selected at random were studied. The age and sex distribution are shown in Figure 2 . These patients sought medical advice at the Division of Community Care Sciences, Dalby, or at a primary health station (OVC) in Lund, Sweden. The study was performed during August and September, 1975. Patients under the age of seven were excluded as it was not practical to draw blood from these young patients at a busy health clinic and it was not justified in view of the diagnosis of the children's condition. Patients with acute otitis media or obvious signs of acute tonsillitis were also excluded. All patients in the study paid their first visit to a doctor for the condition under study and had not taken any antibiotics prior to the visit. The patients were asked to describe their symptoms (Table 8) and to note the day of their onset. A clinical examination was performed and the results recorded. A blood sample was drawn and the patients were instructed to return after two weeks for a second blood specimen to be collected. One hundred and one patients out of 113 returned. A control group was formed of 101 patients who presented to the Centre with symptoms other than those related to the respiratory tract. This control group was matched by sex and age to the patients with respiratory tract infections. One serum sample was drawn from these patients for serological tests. Culture techniques Beta-haemolyticstreptococci. In all the patients with signs and symptoms of respiratory tract infection a culture was taken from the oropharynx. This was obtained with a cotton-tipped swab and inoculated directly onto a blood agar plate which contained 4 % horse and 4 % sheep blood. Isolated streptococci were grouped according to a method described by Christensen et al (8) .

Mycoplasmas. Cultures for mycoplasmas were made from the last 59 patients in the respiratory infection group. The specimens were collected with a cotton-tipped swab from the oro-pharynx and inoculated directly onto three different solid media. Medium "A" consisted of seven parts of heart infusion agar (Difco), two parts of unheated horse serum (v/v), one part of a 25 % (w/v) aqueous fresh yeast extract, and 0.002% (w/v) desoxyribonucleic acid (Sigma). Penicillin G (1000 IU/ml) and thallium acetate (final concentration 1:40,000) were used as bacterial inhibitors. Medium "B'" consisted of the same basic ingredients, but was supplemented with 0.5 mg 2-mercaptoethanol (Sigma) per ml. Medium "C" also had the same basic composition, but to this medium was added 0.5 mg of L-cysteine hydrochloride (Merck) per ml. The agar plates were incubated at 37°C in a humid atmosphere of 90%N2 and 10% CO2. The plates were examined for the presence of colonies at regular intervals by the aid of a stereo-microscope (x120). The plates were discarded after four weeks, if no colonies had been observed. The isolated strains of mycoplasma were identified as to species by growth-inhibition tests using antiserum impregnated paper discs (32).

Sera were stored at -20°C until tested. Complement fixation (CF) tests were performed by a micro-titration technique using disposable plates with the following antigens:

Mycoplasma pneumoniae. The M. pneumoniae antigen was a chloroform-methanol extract (19) of organisms detached from the glass after growth in Roux bottles (21) . This antigen was kindly supplied by Dr. K. Lind, Statens Seruminstitut, Copenhagen.

Viruses. The sera were tested against influenza A and B, para-influenza 1, 2, and 3, as well as against cytomegalovirus (CMV). The antigens were obtained from Statens Bakteriologiska Laboratorium, Solna, Sweden. The sera were also tested against an adenovirus antigen prepared in our own laboratory, and the sera of persons above the age of 60 were tested against a respiratory syncytial (RS) virus antigen (Flow Labs., Ltd.).

Chlamydia. The sera were tested against a Chlamydiapsittaci antigen (Wellcome) for group-specific antibodies to chlamydia.

The sera were assayed for cold agglutinins with a 0.2 % solution of group 0 erythrocytes from adults. Duplicate serum dilutions were made in saline to which was added a 10 % erythroeyte suspension. The readings were made macroscopically after overnight incubation at 4°C.

The Paul-Bunnell test was performed according to the method described by Davidsohn (9) .

so-pharyngitis, virosis, acute bronchitis, acute pharyngitis and conjunctivitis, were represented among the ten most frequent diagnoses. In 1974, respiratory tract infections accounted for about 65 % of all the diagnoses in the children under ten years of age. The corresponding percentage in the age groups 10-19, 20-39, 40-59 and -> 60 years was 45, 32, 18, and 9%.

Beta-haemolytic streptococci were recovered from the oropharynx of ten of the 101 patients studied. Eight of these were group A and two were group G streptococci, while no group C streptococci were found. As indicated, the series did not include patients presenting clinical signs of acute tonsillitis.

Isolation of mycoplasmas M. pneumoniae was recovered from seven patients when the cysteine hydrochloride containing medium (medium "C") was used. The organism was only recovered in one instance when medium "'A", and in two when medium "B" was used. This difference between the three media tested was not observed for other species of mycoplasma (Table 1) .

Viruses. The numbers of cases who had a significant change in the titre of CF antibodies to para-influenza 2 and 3 and to Table 2 viruses are shown in Table 2 . There was no significant difference in the incidence of these antibodies to any of the viruses between the two groups. The titres to influenza A and B in the two groups are shown in Table 3 and those of para-influenza 1, 2, and 3 in Table 4 . The titres of CF antibodies to the other viruses tested, i. e. adeno and CMV, are shown in Table 5 . There is no difference in the distribution of the titres to any of the viruses between the controls and the patients presenting with a respiratory infection (Tables 3, 4 and 5) .

Chlamydia. There was no significant change in the titres of CF antibodies to the chlamydia antigen used in any of the • patients, but one of the controls had a high titre, i. e. 1:80, which was stationary. Mycoplasma pneumoniae. CF antibodies to M. pneumoniae (>--1 : 10) were found in 31% of the patients and in 23 % of the controls. The distribution of the titres in these two groups, which did not disclose any significant difference, is shown in Table 6 . In 7 of the patients, a significant change in the titre was observed ( Table 7 ). The results of the cultures for M. pneumoniae are also included in Table 7 .

Cold agglutinins (-1:16) were found in 20 % of the patients but in none of the controls. In the two patients who had a significant change in the titre of cold agglutinins there was also such a change in the titre of CF antibodies to M. pneumoniae (Table 7) .

In none of the patients or the controls were antibodies found indicating infectious mononucleosis.

The signs and symptoms of the patients are shown in Table  8 . The patients are divided into the various diagnostic categories, e. g. M. pneumoniae, beta-haemolytic streptococci, para-influenza (2 and 3) and adenovirus. The majority of the patients (67%) did not have a definite aetiological diagnosis. Table 2 adenovirus were four, one, and two respectively. No significant change was observed in the titres to the other viruses. The numbers of patients and control subjects who had stationary titres (titre 2 1:10) of CF antibodies to the different 

Respiratory tract infections quantitatively constitute one of the greatest problems in general practice. In the Dalby material about every third patient presented with symptoms and signs of such an infection. The comparatively low frequency of infections associated with group A streptococci in this series must be because patients with definite signs of acute tonsillitis were excluded. It has been estimated that viruses account for 95 % of all cases of acute pharyngitis. Of the remaining 5 %, 95 % are due to beta-haemolytic streptococci group A; group C and G streptococci are implicated in only a few cases. In the present study, group G streptococci were found in two cases, qhe diagnosis of acute pharyngitis had been recorded in all but two of the patients from whom, subsequently, beta-haemolytic streptococci were isolated. Sore throat, fever and headache were the three most frequent symptoms in these patients (Table 8 ).

It is known that viruses are the major causative agents of respiratory tract infections. The test battery used in this study was apparently not particularly suitable to establish the presence of such an infection, since a seroconversion was found in only 7 % of the patients. Rhino and corona viruses which are the most common aetiological agents of respiratory tract infections cannot be routinely diagnosed by serological means as there are no commercially available antigens. Antibodies, at a titre of > 1:10, to all the viruses tested occurred in almost the same proportion in the patients and controls, and the titres had the same distribution. This certainly stresses the fact that it is necessary to examine both acute and convalescent sera. (13), and the Netherlands (16) . The reported incidence of CF antibodies to M. pneumoniae in persons without known respiratory tract infection has been between 5 and 20%. In our control group the corresponding figure was 21%. The present study was performed during the late months of an exceptionally hot summer. There was no M. pneumoniae epidemic during the period of our study. This statement is based bn the low number of positive serological tests for M. pneumoniae that was seen in our laboratory during this period. The laboratory serves an area with about half a million inhabitants in the south of Sweden, including Dalby and Lund.

Most studies of M. pneumoniae infections outside hospitals have been obtained from studies conducted in closed communities, such as military units, university campuses (7, 10) and in families (12) . It has been suggested that M. pneumoniae more often causes upper respiratory tract infections than pneumonia, but figures supporting this assumption are few. The reported incidence of M. pneumoniae in some selected series of patients with upper respiratory tract infections are shown in Table 9 . Incidence figures of M. pneumoniae infections in randomly selected patients in general practice have not to our knowledge been reported. We found evidence of a M. pneumoniae infection in 16 patients; a significant rise in titre of CF antibodies was found in seven, while four had high, but stationary, titres. The epidemiological data and the history strongly suggested an infection with M. pneumoniae in these four patients. Cold agglutinins were found in two of them. In five patients there was no serological evidence of a M. pneumoniae infection, but the organism was recovered from the oro-pharynx. CF, indirect haemagglutination (IHA) and immunofluorescence (IF) tests have been used to detect serum antibodies Infection 4 (1976) Suppl. 1 S (6) Marine recruits (1961) Grayston (15) Miscellaneous groups (1965) of patients Evans (19) University students (1967) Hers (16) Selected groups of (1967) civilians (studied between 1961 and 1966) Glezen (14) Patients in pediatric (1967 The frequency with which M. pneumoniae has been isolated from patients with serological evidence of infection with this organism has been low in most studies. Body fluids contain mycoplasmacidal substances that might influence the recovery rate (27) . The composition of the culture media in general use may not be optimal for the isolation ofM. pneumoniae from clinical specimens, although they may support the growth of "'laboratory strains" of M. pneumoniae. In the present study we found that the addition of a reducing substance, i. e. cysteine hydrochloride, gave a higher recovery rate, but this observation needs further evaluation. Cysteine hydrochloride has been used earlier in culture media for ureaplasmas (30) . It should be noted that the isolation rate of more rapidly growing mycoplasma species, e. g. M. hominis, M. orale, and M. salivarium, was not influenced by the presence of this substance. It is notable that M. pneumoniae was isolated from five patients who did not have any se-rological evidence of such an infection, which indicates the role of culture studies in attempts to establish upper respiratory tract infections with this organism. Cold agglutinins are known to occur in 50 to 75 % of patients with signs of pneumonia and with a significant change in the titre of CF antibodies to M. pneumoniae. In our series of patients with non-pneumonic respiratory tract infections, cold agglutinins were found in three of those seven with such a titre change. It has been recently demonstrated that cold agglutinins may also occur in CMV infections of the respiratory tract (24), but we did not find any evidence of such an infection in our cold agglutinin positive cases. The present study seems to indicate that M. pneumoniae infections are not uncommon in patients presenting with symptoms of upper respiratory tract infection. This type of infection was the most common diagnosis made in our series. But it must be stressed that children below the age of seven and cases of acute tonsillitis and acute otitis media were excluded. The physician often obtains little guidance from the history, and the clinical examination of the patient does not often al- S low the physician to establish the aetiology of a respiratory tract infection. The signs and symptoms in the present series of patients with either a streptococcal, viral or M. pneumoniae infection showed certain common characteristics, but the clinical picture was not sufficiently distinct to make a definite diagnosis without the laboratory tests (Table 8 ). The information the physician obtains from the microbiological laboratory often arrives too late to be of help in the treatment of the individual case. The physician therefore usually has to decide whether or not to treat a respiratory tract infection without the knowledge of the aetiology of the patient's condition. A retrospective study of the treatment which had been giveia to our patients clearly indicates the difficulties involved (Table 10 ). The liberal prescription of penicillin to patients presenting with signs of pharyngo-tonsillitis may seem justified in relation to the difficulty of basing the diagnosis of an infection with beta-haemolytic streptococci group A on clinical grounds (1, 4) , and when it is known that early institution of penicillin reduces the incidence of complications, such as acute glomerulonephritis (18) . Such a therapy may also be of epidemiological significance.

The duration of fever and cough in cases of pneumonia caused by M. pneumoniae may be reduced by the use of tetracyclines or erythromycin. Whether such treatment of upper respiratory tract infections caused by this organism is of any benefit is not known, and it could not be determined from our study. M. pneumoniae is susceptible to these antibiotics in vitro (25), but they have been shown to persist in vivo despite their usage (31) . Resumen: Se estudi6 la incidencia de infecciones respiratorias en pacientes que visitaban al mddico en un centro asistencial (Dalby) durante 1973 y 1974. Uno de cada tres pacientes de los examinados en aquel centro sanitario presentaba signos de la mencionada patologia. En estos grupos de edades: < 10, 10-19, 20-29, 40-59 y _> 60 alms, las infecciones del tracto respiratorio representaron el 65 %, 45 %, 32%, 18% y el 9 % respectivamente del nfimero total de dia-gn6sticos realizados durante el afio 1974. Se estudi6 la etiologia de las infecciones agudas del firbol respiratorio en una serie de pacientes examinados en este centro sanitario. La serie incluy6 casos seleccionados al azar, excluyendo nifms menores de 7 afios de edad asf como pacientes que presentaban sfntomas de otitis media aguda y amigdalitis. Se intent6 establecer la etiologfa basfindose en la historia del proceso, el examen clfnico y el examen bacteriol6gico con cultivo para determinar la presencia del estreptococo beta-hemolf- pneumoniae nel 16%, con Io streptococco beta-emoicio nel 9% e con virus (adeno e para influenza) nel 7 %. La presente communicazione sottolinea il ruolo del M. pneumoniae nelle infezioni dell'apparato respiratorio superiore, poich~ pochi dati sono disponibili su queste infezioni nei pazienti osservati nella pratica generica. Viene anche discussa la difficolt~ di stabilire Feziologia delle infezioni dell'apparato respiratorio e i conseguenti dubbi sul trattamento.

Acute pharyngeal infection, aetiological and diagnostic viewpoints

Antibodies to brain and other tissues in cases of Mycoplasmapneumoniae infection

Mycoplasma pneumoniae infection in hospitalized patients with acute respiratory illness. Acta path. microbiol, scand

The accuracy of diagnosis of beta streptococcal infections on clinical grounds

Aspects of the diagnosis of throat infections with beta haemolytic streptococci group A

Eaton agent pneumonia

Epidemiology of Mycoplasma pneumoniae infection in military recruits

New method for the serological grouping of streptococci with specific antibodies adsorbed to protein A-containing staphylococci

Serologic diagnosis of infectious mononucleosis

Mycoplasma pneumoniae infections in University of Wisconsin students

Pneumonia, Mycoplasma pneumoniae

Epidemiology of Mycoplasma pneumoniae infection in families

Mycoplasma pneumoniae pneumonia in an urban area. Five years of surveillance

Group A streptococci, mycoplasmas and viruses associated with acute pharyngitis

Mycoplasma pneumoniae infections

Infection with Mycoplasma pneumoniae in civilians in the Netherlands

Acute respiratory virus infections

Prevention of acute glomerulonephritis with early treatment of tonsillitis with penicillin. Scand

Eaton pleuropneumonia-like organism (Mycoplasma pneumoniae) complement-fixing antigen: Extraction with organic solvents

An indirect haemagglutination test for serum antibodies against Mycoplasma pneumoniae using formalinized, tanned sheep erythrocytes. Acta path. microbiol, scand

Incidence of Mycoplasma pneumoniae infection in Denmark from 1958 to 1969. Acta path. microbiol, scan& Sect

Cold agglutinin production and cytomegalovirus infection. Scand

Human respiratory tract infections with mycoplasmas and their in vitro susceptibility to tetracyclines and some other antibiotics

Mycoplasma pneumoniae infection: A cause of acute pancreatitis and non-specific, reactive hepatitis in man and experimentally infected animals

New approaches to the isolation of mycoplasmas

Persistent cerebellar symptoms after infection with Mycoplasma pneumoniae. Scand

Mycoplasma pneumoniae infection in the United Kingdom-1967-73

Ureaplasma urealyticure gen. nov., sp. nov.: proposed nomenclature for the human T (T-strain) mycoplasmas

Growth inhibition test for identification of Mycoplasma species utilizing dried antiserum-impregnated paper discs