Journal
of Hospital
Infection
(1992)
21, 213-221
Outbreak of Mycoplasma pneumoniae infection among hospital personnel studied by a nucleic acid hybridization test M. Kleemola” “National
Public
Health University
and C. Jokinen-/-
Institute, Hospital,
00300 Helsinki, Finland 70210 Kuopio, Finland
Accepted for publication
3 April
and TKuopio
1992
Summary: An outbreak of ~yroplnsmapneumoniae (MP) infection occurred during the period March-May 1989 among the personnel of the Accident and Emergency Department of the Kuopio University Hospital, Kuopio, Finland. The index patient was a young male orderly, who fell ill with severe pneumonia. His tracheal mucus sample proved to be strongly positive for MI’ when tested by a commercial DNA-RNA hybridization test (Gen-Probe). After the index patient two additional staff members (an orderly and a nurse) fell ill with pneumonia and 66 others showed symptoms of upper respiratory infection or fever. The most frequent symptoms were a sore throat, a cough, r-hinitis and headaches. All 97 employees of the department were tested for the presence of MP in April-May 1989 using throat swabs as test material. Forty-three (44%) were found to be positive for MP by the ‘Gen-Probe’ test. Eight (19%) of the MP positive staff were completely asymptomatic. The MP positive staff were retested about 3 weeks later, whereupon 40 (93%) had become negative. Most of the persons involved in this outbreak suffered only suggesting that MP outbreaks like the from mild respiratory symptoms, present one may easily pass unnoticed. Keywords: Hospital-acquired; nucleic acid hybridization;
hospital Outbreak.
personnel;
Mycoplasma
pneumoniae;
Introduction Mycoplasma
pneumoniae (MP) is a common respiratory pathogen occurring endemically in different populations with epidemic outbreaks at irregular intervals.“2,3 Pneumonia is an important manifestation of MP infection, although it only occurs in a small proportion of all those affected. Milder forms of respiratory infection, such as coughs, sore throats and rhinitis, are much more commorq4,’ however, MP is still responsible for approximately 15 % of pneumonias in the general population.* L)uring epidemic peaks, and in closed communities such as military garrisons and student dormitories, the attack rate of MP pneumonia may be as high as 5OoA.5,6 Transmission of Correspondence to: Dr Marjaana SF-00300 Helsinki, Finland. 0195-67fl1/92/070213+09
Kleemola,
National
$03 00/O
213
Public Health Institute,
Mannerheimintie
166,
214
M. Kleemola
and
C. Jokinen
the infection seems to require strong exposure and close contact.’ An outbreak among hospital personnel has been previously described by Fisher et ~1.~ and among hospitalized patients by Kashiwagi et aZ.9 The present outbreak was recognized when three of the employees working in the Accident and Emergency Department (AED) of the Kuopio University Hospital and an ambulance driver bringing patients to the AED, fell ill with pneumonia between 4 and 19 April 1989. The tracheal mucus sample of the index patient, who required intensive care, was strongly positive by the MP hybridization test (Gen-Probe). By 28 April a total of 15 symptomatic persons working in the AED and three patients admitted to other departments had been found to be positive for MP. As we suspected that this might be the start of a large outbreak of MP infection, we decided to examine all personnel of the AED for symptoms of respiratory infection and for the presence of MP. The newly developed ‘Gen-Probe’ test was chosen as a test method because of its convenience and rapidity. Subjects
and methods
Description of the department and the subjects The Kuopio University Hospital is an 828-bed reference hospital serving a population of 250 000 in Eastern Finland. The emergency room and duty ward, where patients are observed overnight, together form the AED. The same personnel work in both units and will usually have some degree of contact with each other on most workdays. The 97 people working in the AED during 1 April to 31 May 1989, either permanently or temporarily, were all included in the study. Seventy-four were female and 23 male. The median age of the staff was 32 years (range 21-55 years). Collection of data and samples Two similar questionnaires were completed by the subjects (at the beginning and end of May; Figure 1). The questions were concerned with age, sex, occupation, family size, occurrence of respiratory symptoms during March-May, antimicrobial therapy and sick leave. In April-May (with a peak in the first week of May; Figure 1) throat swabs were obtained from 96 persons and tracheal mucus from one. The throat specimens were taken from the posterior pharynx with a cotton swab that was subsequently immersed in 1.5 ml of Hank’s balanced salt solution containing 0.5% of gelatin and sent refrigerated to the laboratory where the test was performed the following day. The laboratory results were reported by telephone to the AED on the day of testing and confirmed later by a written reply. The results were used for therapeutic decisions, as well as for planning the subsequent individual sampling. If the first MP result was positive, the sampling was repeated at 3-week intervals until the result was negative (Figure 1). The personnel were informed of the aims of the study and the
Mycoplasma
pneumoniae
outbreak
215
-
*
*I-*
-*-F
13 /
I
14
/
15 April
28 Weeks
16 I
May
Figure 1. Timing of the samples for the ‘Gen-Probe’ test and completion of the questionnaires during an outbreak of Mycoplasma pneumoniae infection among the personnel of the Accident and Emergency Department of the Kuopio University Hospital in March-May, 1989. Two similar questionnaires were completed during the study: at the beginning and the end of May. Dark bar = completion of the first questionnaire (N= 96), light bar = completion of the second questionnaire (N= 97), * = the first sample (N= 97), 0 = the second sample (IV= 43), X =the third sample (N= 3).
voluntariness of participation orally as well as on the questionnaires, which they completed and signed. The study protocol was approved by the Ethics Committee of the University of Kuopio and the Kuopio University Hospital. Microbiological methods The ‘Gen-Probe’ Rapid Diagnostic System (Gen-Probe, San Diego, CA, USA) for diagnosis of MP infection had been introduced a couple of years earlier. It is based on hybridization in solution where a ‘251-labelled DNA probe hybridizes with a sequence of the ribosomal RNA of MP. The whole procedure only takes about 2 h. The test was performed according to the instructions supplied with the kit. Briefly, the sample was centrifuged at 12 000 gfor 10 min, the supernatant was removed and the RNA of the cells liberated using a lysing solution. 150 ~1 of the probe (‘251-labelled DNA) was added and the hybridization was allowed to take place at + 72°C for 1 h.
M. Kleemola
216
and C. Jokinen
The formed hybrids were adsorbed to a separation suspension at + 72°C for 5 min and, after washing, the radioactivity of the pellet was measured for 1 min in a gamma counter. The result was expressed as a ratio of the sample counts to the counts of a negative control, after subtracting the background from both. A ratio of 3.0 or greater was interpreted as positive. The ‘Gen-Probe’ test is claimed to be specific for MP with one exception, i.e. M. genitalium. This species, originally isolated from the urogenital area, lo has later been found, together with MP, in respiratory secretions of patients with acute respiratory infections, l1 but its possible causative role in such cases is unclear. The sensitivity of the ‘Gen-Probe’ test varies from 22-89%,12-” or even lOO%, l6 depending on the reference method, and also on the type and quality of the specimens. In our laboratory the ‘Gen-Probe’ test has detected only 30% of serologically positive cases when throat swabs were used as test material, whereas the respective figure was 75% for sputum samples.13 In the present study only throat swabs were possible, as most of the subjects did not produce sputum. The tracheal mucus sample from the index patient was cultured for MP as described previously,‘3 and MP serology was performed on his paired serum samples using a standard CF test.17 Statistical methods Chi-square, Student’s statistical analyses.
t-test
Description
and Fisher’s
of the outbreak
exact
test were
used for the
and results
Clinical data The index patient was a 29-year-old previously healthy male orderly who fell ill on 30 March with cough, fever and muscle pains. He developed infiltrates, left-sided pleural effusion and bilateral rales, radiological respiratory failure. He was admitted to the Intensive Care Unit on the seventh day of illness, managed without assisted ventilation, transferred to the ward on day 10 and discharged on day 15. MP infection was diagnosed by the ‘Gen-Probe’ test from his tracheal mucus collected on admission, and it was later confirmed by a positive MP culture from the same sample, as well as by a 32-fold increase in specific complement fixing antibodies of paired sera. The patient received erythromycin and cefuroxime intravenously for 9 days, followed by oral erythromycin for 7 days. He returned to work on day 40. Another male orderly and a female nurse also had radiologically confirmed pneumonia, but their ‘Gen-Probe’ tests were negative. No other severe infections occurred. Overall, 69 of the subjects (71 “A) had symptoms and/or findings generally known to be associated with MP infection’ during March-May. No difference in the occurrence of such symptoms and findings were found by
Mycoplasma
pneumoniae
outbreak
217
age, sex or occupational group. The most frequent symptoms (Table I) were sore throat (occurring in 62% of all symptomatic subjects), cough (54%) and rhinitis (49%), while only 25% had fever (axillary temperature > 37~5°C). About half of the illnesses began in the last 2 weeks of April (Figure 2). Number of persons positive for MP The ‘Gen-Probe’ test was positive at least once in 43 persons including 5 1% of the 69 with and 29% of the 28 without symptoms (P~0.05). Persons with sore throat were significantly more often positive than those without (PC 0.01). As to the other symptoms and findings, the differences between MP positive and negative subjects were not statistically significant (Table I). Eight (19’/) of MP positive persons remained completely asymptomatic. Treatment A treatment schedule was planned with the occupational health physician of the hospital, who saw most of the subjects with medical problems and had access to the MP results. At the beginning of the study we recommended erythromycin or doxycycline and sick leave to all persons with symptoms suggestive of MP infection, or who were asymptomatic but MP positive. As most infections later appeared to be mild or asymptomatic, and as systematic sick leave would have resulted in collapse of the medical service, these recommendations were later restricted to those persons who were
Table
I. Correlation
of clinical symptoms and signs in persons during Mycoplasma pneumoniae hybridization test Mycoplasma test Mycoplasma test positive (N= 43) negative (N= 54) N N % %
Symptoms Cough Rhinitis Sore throat Headache Other* Signs Axillary temperature > 37.5”C Pneumonia7 Other1 One or more of the above:
Total
(N=
the outbreak 97)
N
%
with
Significance
26 12 6
44 35 60 28 14
18 19 17 12 9
33 35 31 22 15-
37 34 43 24 15
38 35 44 25 15
NS NS PC O-01 NS NS
9 1 4
21 2 9
8 2 2
15 4 4
17 3 6
18 3 6
NS NS NS
35
81
34
63
69
71
NS = not significant. * Rash (three subjects), muscle pain (4), chest pain (2), dyspnoea (5), lassitude t Radiologically confirmed cases. $ Sinusitis (l), lymphadenitis (3), cyanosis (I), otitis media (1).
for longer
Pi
0.05
than 1 week (1).
M. Kleemola
and
C. Jokinen
r 17
’
18
23
Weeks
I April Marcll May I I Figure 2. Weekly numbers of persons falling ill with acute respiratory infection, and time of the first positive samples during an outbreak of Mycoplasma pneumoniae (MP) infection among the personnel of the Accident and Emergency Department of the Kuopio University Hospital in March-May 1989. Solid line=symptomatic MP-positive persons: their first positive samples (N= 35), dark bar = symptomatic MP-positive persons: onset of respiratory symptoms (the same persons as above; N=35), dotted line=asymptomatic MP-positive persons: their first positive samples (N= S), light bar = symptomatic MP-negative persons: onset of respiratory symptoms (N= 34).
definitely ill. Thirty subjects (31%) eventually received erythromycin and five (5%) doxycycline for 6-l 8 days. Three (3%) received other antibiotics for tonsillitis, wound or urinary tract infection. Twenty-seven persons (28%) were on sick leave for 2-39 d because of acute respiratory infection, fever, or asymptomatic shedding of MP. Only the index patient and one of the two other patients with pneumonia were admitted to hospital.
Duration
of the symptoms and MP shedding
The mean duration in days of the various symptoms was: cough 11.6, rhinitis 14.9, sore throat 9.9, fever 3.8, headache 4.9 and other symptoms 7.0. The duration of these symptoms was very similar irrespective of the MP results or whether antibiotics were given or not. The second sample, taken from the 43 MP-positive persons approximately 23 days after the first, was negative in 40 cases. The three
Mycoplasma
pneumoniae
remaining positives were sampled a third negative. The longest time a subject remained 48 and 57 days. Source
of infection
and transmission
outbreak
219
time, whereupon all were positive for MP was between
in families
The personnel themselves were not aware of any obvious source of infection. The occurrence of symptoms was unrelated to household size: symptomatic employees had an average of 1.6 family contacts and asymptomatic ones, 1.5. There was, however, a difference in the occurrence of clinical illness between the families of symptomatic and asymptomatic individuals. Acute respiratory tract infection or fever during March-May was reported for 53% of the 113 household contacts of symptomatic employees, half of them (30) falling ill before the employees and half (30) later. In the families of asymptomatic persons, only 29% of the 42 family members reported illness (PC 0.01). Some 47% of the family contacts of 43 MP-positive subjects and 46% of those of 54 MP-negative subjects had symptoms, and the average family size was similar in the two groups. Discussion
This study investigated an outbreak of MP infection which occurred among hospital staff. Most of the infections were mild: of 69 symptomatic subjects only three (4%) had pneumonia and 17 (25%) had fever (> 37.5”(Z). In addition, eight (19’/,) of the MP-positive subjects were completely asymptomatic. These figures are in general agreement with published data 4,5,18,19 and further emphasize the mildness of most cases of this disease. If it’had not been for the one severe case in the beginning, the outbreak might well have escaped attention. One aim of the present study was to examine the usefulness of the recently introduced rapid ‘Gen-Probe’ test in monitoring an outbreak of MP infections. Even though a sensitivity as low as 30% with throat swabs has been previously described in our laboratory,13 50% of the symptomatic persons proved to be MP positive. It is thus concluded that, although for individual cases only a positive test is predictive, the ‘Gen-Probe’ test can be used for the detection of an outbreak and for following its course. It is probable that the infections were both community and hospital acquired. A community-acquired origin was suggested by the large number of family contacts whose symptoms started before those of the staff and additional hospital spread indicated by the higher percentage of symptomatic individuals among the staff (70%) compared to family members (45%). Given the low sensitivity of the ‘Gen-Probe’ test using throat swabs, it is possible that within a short time period (2 months) up to 90% of the personnel were infected with MP and 70% had symptoms. Given such an extensive outbreak it is likely that transmission to patients would also have occurred. No evidence of this was obtained; however, such spread may have
220
M. Kleemola
and C, Jokinen
been missed. The patients remained in the hospital for an average of only 5 days, i.e. much less than the incubation period of MP infection” and given the general mildness of the infections, their development following discharge might not have been deemed important. It is also possible that MP infection arising in patients with or convalescing from severe diseases might have resulted in serious manifestations falsely ascribed to the underlying disease alone, How could one prevent the spread of infection among hospital staff and patients? In the light of the present experience this seems a very difficult task. Insisting on sick leave for affected staff might be effective; however, such a course of action is complicated by (a) the many MP positives among asymptomatic individuals, (b) the low sensitivity of the ‘Gen-Probe’ test, and finally (c) the very large proportion of staff who might be infected. For maximum safety the whole ward would need to be closed-but the consequences to patient care could be more deleterious than exposure to MP. Neither the limited sick leave practised in the current episode nor the instigation of antibiotic therapy was successful in stopping the spread of infection” A further difficulty in controlling outbreaks like this is posed by their insidious beginning-when would one decide to institute major control measures? One possibility would be continuous monitoring of all (even mild) respiratory infections in the staff using the ‘Gen-Probe’ test. The first positive finding might then be a signal for more extensive probing of the whole personnel, and for ordering sick leave if several positives are found. This raises several questions: should other respiratory pathogens be monitored in the same way and would this scheme be cost effective? In which wards should it be instituted? These are questions that remain open with present knowledge and call for further investigation. Such studies should also examine the extent of transmission of these infections to patients and their impact on disease and recovery. We thank the personnel of the Accident and Emergency Department of Kuopio University Hospital for their invaluable co-operation, and MS Sari Jokinen as well as other technicians in the laboratory of the National Public Health Institute for their skilful aid in testing the samples. We are especially grateful to Prof. P. Helena Makela for critical evaluation of our manuscript. Our thanks are also due to Mr Jukka Lauronen for the statistical analyses and MS Riitta Brelih for revising our manuscript.
References 1. Denny FW, Clyde WA, Glezen WP. Mycoplasmapneumoniae disease: clinical spectrum, pathophysiology, epidemiology, and control. r Infect Dis 1971; 123: 74-92. 2. Foy HM, Kenny GE, Cooney MK, Allan ID. Long-term epidemiology of infections with Mycoplasma pneumoniae. J Infect Dis 1979; 139: 681-687. 3. Lind K, Bentzon MW. Changes in the epidemiological pattern of Mycoplasma pneumoniae infections in Denmark. A 30 years survey. Epidemiol Infect 1988; 101: 377-386. 4. Smith, Chanock RM, Friedewald WT, Alford RH. Mycoplasma pneumoniae infections in volunteers. Ann NY Acad Sci 1967; 143: 471483.
Mycoplasma
pneumoniae
outbreak
221
5. Broughton R. Infections due to Mycoplasma pneumoniae in childhood. Pediatr Infect Dis J 1986; 5: 71-85. 6. Mogabgab W. Mycoplasma pneumoniae and adenovirus respiratory illnesses in military and university personnel, 1959-1966. Am Rev Respir Dis 1968; 97: 345-358. 7. Steinberg P, White RJ, Fuld SL, Gutekunst RR, Chanock RM, Senterfit LB. Ecology of Mycoplasmapneumoniae infections in marine recruits at Parris Island, South Carolina. Am r Epidemiol 1969; 89: 62273. 8. Fisher B, Yu B, Armstrong D, Magi11 J. Outbreak of Mycoplasma pneumoniae infection among hospital personnel. Am g Med Sci 1978; 276: 205-209. 9. Kashiwagi S, Hayashi J, Nomura H et al. An outbreak of Mycoplasma pneumoniae infections in a hospital in Japan. Kurume Med J 1985; 32: 2933296. 10. Tully JG, Taylor-Robinson D, Cole RM, Rose DL. A newly discovered mycoplasma in the human urogenital tract. Lancet 1981; i: 1288-1291. 11. Baseman JB, Dallo SF, Tully JG, R ose DL. Isolation and characterization of Mycoplasma genital&m strains from the human respiratory tract. J Clin Microbial 1988; 26: 226662269. 12. Harris R, Marmion BP, Varkanis G, Kok T, Lunn B, Martin J. Laboratory diagnosis of Mycoplasmapneumoniae infection. 2. Comparison of methods for the direct detection of specific antigen or nucleic acid sequences in respiratory exudates. Epidemiol Infect 1988;
101: 685-694. 13. Kleemola pneumoniae
SRM, Karjalainen infection: clinical
JE, RCty RKH. Rapid diagnosis of Mycoplasma evaluation of a commercial probe test. 2 Infect Dis 1990;
162: 70-75. 14. Hata D, Kuze F, Mochizuki Y et al. Evaluation of DNA probe test for rapid diagnosis of Mycoplasma pneumoniae infections. J Pediatr 1990; 116: 273-276. 15. Dular R, Kajioka R, Kasatiya S. Comparison of Gen-Probe commercial kit and culture technique for the diagnosis of Mycoplasma pneumoniae infection. r Clin Microbial 1988; 26: 106881069. 16. Tilton RC, Dias F, Kidd H, Ryan RW. DNA probe versus culture for detection of Mycoplasma pneumoniae in clinical specimens. Diagn Microbial Infect Dis 1988; 10: 1099112. 17. Hawkes RA. General principles underlying laboratory diagnosis of viral infections. In: Lennette EH, Schmidt NJ, Eds. Diagnostic Procedures for Viral Rickettsial and Chlamydial Infections. 5th edn. Washington, DC: American Public Health Association 1979: 3542. 18. Biberfeld G, Sterner G. A study of Mycoplasmapneumoniae infections in families. Stand .r Infect Dis 1969; 1: 39-46. 19. Grayston JT, Kenny GE, Foy HM., Kronmal RA, Alexander ER. Epidemiological studies of Mvcoalasma Pneumoniae infections in civilians. Ann NY Acad Sci 1967: 143: 436446. - _ _
of Hospital
Infection
(1992)
21, 213-221
Outbreak of Mycoplasma pneumoniae infection among hospital personnel studied by a nucleic acid hybridization test M. Kleemola” “National
Public
Health University
and C. Jokinen-/-
Institute, Hospital,
00300 Helsinki, Finland 70210 Kuopio, Finland
Accepted for publication
3 April
and TKuopio
1992
Summary: An outbreak of ~yroplnsmapneumoniae (MP) infection occurred during the period March-May 1989 among the personnel of the Accident and Emergency Department of the Kuopio University Hospital, Kuopio, Finland. The index patient was a young male orderly, who fell ill with severe pneumonia. His tracheal mucus sample proved to be strongly positive for MI’ when tested by a commercial DNA-RNA hybridization test (Gen-Probe). After the index patient two additional staff members (an orderly and a nurse) fell ill with pneumonia and 66 others showed symptoms of upper respiratory infection or fever. The most frequent symptoms were a sore throat, a cough, r-hinitis and headaches. All 97 employees of the department were tested for the presence of MP in April-May 1989 using throat swabs as test material. Forty-three (44%) were found to be positive for MP by the ‘Gen-Probe’ test. Eight (19%) of the MP positive staff were completely asymptomatic. The MP positive staff were retested about 3 weeks later, whereupon 40 (93%) had become negative. Most of the persons involved in this outbreak suffered only suggesting that MP outbreaks like the from mild respiratory symptoms, present one may easily pass unnoticed. Keywords: Hospital-acquired; nucleic acid hybridization;
hospital Outbreak.
personnel;
Mycoplasma
pneumoniae;
Introduction Mycoplasma
pneumoniae (MP) is a common respiratory pathogen occurring endemically in different populations with epidemic outbreaks at irregular intervals.“2,3 Pneumonia is an important manifestation of MP infection, although it only occurs in a small proportion of all those affected. Milder forms of respiratory infection, such as coughs, sore throats and rhinitis, are much more commorq4,’ however, MP is still responsible for approximately 15 % of pneumonias in the general population.* L)uring epidemic peaks, and in closed communities such as military garrisons and student dormitories, the attack rate of MP pneumonia may be as high as 5OoA.5,6 Transmission of Correspondence to: Dr Marjaana SF-00300 Helsinki, Finland. 0195-67fl1/92/070213+09
Kleemola,
National
$03 00/O
213
Public Health Institute,
Mannerheimintie
166,
214
M. Kleemola
and
C. Jokinen
the infection seems to require strong exposure and close contact.’ An outbreak among hospital personnel has been previously described by Fisher et ~1.~ and among hospitalized patients by Kashiwagi et aZ.9 The present outbreak was recognized when three of the employees working in the Accident and Emergency Department (AED) of the Kuopio University Hospital and an ambulance driver bringing patients to the AED, fell ill with pneumonia between 4 and 19 April 1989. The tracheal mucus sample of the index patient, who required intensive care, was strongly positive by the MP hybridization test (Gen-Probe). By 28 April a total of 15 symptomatic persons working in the AED and three patients admitted to other departments had been found to be positive for MP. As we suspected that this might be the start of a large outbreak of MP infection, we decided to examine all personnel of the AED for symptoms of respiratory infection and for the presence of MP. The newly developed ‘Gen-Probe’ test was chosen as a test method because of its convenience and rapidity. Subjects
and methods
Description of the department and the subjects The Kuopio University Hospital is an 828-bed reference hospital serving a population of 250 000 in Eastern Finland. The emergency room and duty ward, where patients are observed overnight, together form the AED. The same personnel work in both units and will usually have some degree of contact with each other on most workdays. The 97 people working in the AED during 1 April to 31 May 1989, either permanently or temporarily, were all included in the study. Seventy-four were female and 23 male. The median age of the staff was 32 years (range 21-55 years). Collection of data and samples Two similar questionnaires were completed by the subjects (at the beginning and end of May; Figure 1). The questions were concerned with age, sex, occupation, family size, occurrence of respiratory symptoms during March-May, antimicrobial therapy and sick leave. In April-May (with a peak in the first week of May; Figure 1) throat swabs were obtained from 96 persons and tracheal mucus from one. The throat specimens were taken from the posterior pharynx with a cotton swab that was subsequently immersed in 1.5 ml of Hank’s balanced salt solution containing 0.5% of gelatin and sent refrigerated to the laboratory where the test was performed the following day. The laboratory results were reported by telephone to the AED on the day of testing and confirmed later by a written reply. The results were used for therapeutic decisions, as well as for planning the subsequent individual sampling. If the first MP result was positive, the sampling was repeated at 3-week intervals until the result was negative (Figure 1). The personnel were informed of the aims of the study and the
Mycoplasma
pneumoniae
outbreak
215
-
*
*I-*
-*-F
13 /
I
14
/
15 April
28 Weeks
16 I
May
Figure 1. Timing of the samples for the ‘Gen-Probe’ test and completion of the questionnaires during an outbreak of Mycoplasma pneumoniae infection among the personnel of the Accident and Emergency Department of the Kuopio University Hospital in March-May, 1989. Two similar questionnaires were completed during the study: at the beginning and the end of May. Dark bar = completion of the first questionnaire (N= 96), light bar = completion of the second questionnaire (N= 97), * = the first sample (N= 97), 0 = the second sample (IV= 43), X =the third sample (N= 3).
voluntariness of participation orally as well as on the questionnaires, which they completed and signed. The study protocol was approved by the Ethics Committee of the University of Kuopio and the Kuopio University Hospital. Microbiological methods The ‘Gen-Probe’ Rapid Diagnostic System (Gen-Probe, San Diego, CA, USA) for diagnosis of MP infection had been introduced a couple of years earlier. It is based on hybridization in solution where a ‘251-labelled DNA probe hybridizes with a sequence of the ribosomal RNA of MP. The whole procedure only takes about 2 h. The test was performed according to the instructions supplied with the kit. Briefly, the sample was centrifuged at 12 000 gfor 10 min, the supernatant was removed and the RNA of the cells liberated using a lysing solution. 150 ~1 of the probe (‘251-labelled DNA) was added and the hybridization was allowed to take place at + 72°C for 1 h.
M. Kleemola
216
and C. Jokinen
The formed hybrids were adsorbed to a separation suspension at + 72°C for 5 min and, after washing, the radioactivity of the pellet was measured for 1 min in a gamma counter. The result was expressed as a ratio of the sample counts to the counts of a negative control, after subtracting the background from both. A ratio of 3.0 or greater was interpreted as positive. The ‘Gen-Probe’ test is claimed to be specific for MP with one exception, i.e. M. genitalium. This species, originally isolated from the urogenital area, lo has later been found, together with MP, in respiratory secretions of patients with acute respiratory infections, l1 but its possible causative role in such cases is unclear. The sensitivity of the ‘Gen-Probe’ test varies from 22-89%,12-” or even lOO%, l6 depending on the reference method, and also on the type and quality of the specimens. In our laboratory the ‘Gen-Probe’ test has detected only 30% of serologically positive cases when throat swabs were used as test material, whereas the respective figure was 75% for sputum samples.13 In the present study only throat swabs were possible, as most of the subjects did not produce sputum. The tracheal mucus sample from the index patient was cultured for MP as described previously,‘3 and MP serology was performed on his paired serum samples using a standard CF test.17 Statistical methods Chi-square, Student’s statistical analyses.
t-test
Description
and Fisher’s
of the outbreak
exact
test were
used for the
and results
Clinical data The index patient was a 29-year-old previously healthy male orderly who fell ill on 30 March with cough, fever and muscle pains. He developed infiltrates, left-sided pleural effusion and bilateral rales, radiological respiratory failure. He was admitted to the Intensive Care Unit on the seventh day of illness, managed without assisted ventilation, transferred to the ward on day 10 and discharged on day 15. MP infection was diagnosed by the ‘Gen-Probe’ test from his tracheal mucus collected on admission, and it was later confirmed by a positive MP culture from the same sample, as well as by a 32-fold increase in specific complement fixing antibodies of paired sera. The patient received erythromycin and cefuroxime intravenously for 9 days, followed by oral erythromycin for 7 days. He returned to work on day 40. Another male orderly and a female nurse also had radiologically confirmed pneumonia, but their ‘Gen-Probe’ tests were negative. No other severe infections occurred. Overall, 69 of the subjects (71 “A) had symptoms and/or findings generally known to be associated with MP infection’ during March-May. No difference in the occurrence of such symptoms and findings were found by
Mycoplasma
pneumoniae
outbreak
217
age, sex or occupational group. The most frequent symptoms (Table I) were sore throat (occurring in 62% of all symptomatic subjects), cough (54%) and rhinitis (49%), while only 25% had fever (axillary temperature > 37~5°C). About half of the illnesses began in the last 2 weeks of April (Figure 2). Number of persons positive for MP The ‘Gen-Probe’ test was positive at least once in 43 persons including 5 1% of the 69 with and 29% of the 28 without symptoms (P~0.05). Persons with sore throat were significantly more often positive than those without (PC 0.01). As to the other symptoms and findings, the differences between MP positive and negative subjects were not statistically significant (Table I). Eight (19’/) of MP positive persons remained completely asymptomatic. Treatment A treatment schedule was planned with the occupational health physician of the hospital, who saw most of the subjects with medical problems and had access to the MP results. At the beginning of the study we recommended erythromycin or doxycycline and sick leave to all persons with symptoms suggestive of MP infection, or who were asymptomatic but MP positive. As most infections later appeared to be mild or asymptomatic, and as systematic sick leave would have resulted in collapse of the medical service, these recommendations were later restricted to those persons who were
Table
I. Correlation
of clinical symptoms and signs in persons during Mycoplasma pneumoniae hybridization test Mycoplasma test Mycoplasma test positive (N= 43) negative (N= 54) N N % %
Symptoms Cough Rhinitis Sore throat Headache Other* Signs Axillary temperature > 37.5”C Pneumonia7 Other1 One or more of the above:
Total
(N=
the outbreak 97)
N
%
with
Significance
26 12 6
44 35 60 28 14
18 19 17 12 9
33 35 31 22 15-
37 34 43 24 15
38 35 44 25 15
NS NS PC O-01 NS NS
9 1 4
21 2 9
8 2 2
15 4 4
17 3 6
18 3 6
NS NS NS
35
81
34
63
69
71
NS = not significant. * Rash (three subjects), muscle pain (4), chest pain (2), dyspnoea (5), lassitude t Radiologically confirmed cases. $ Sinusitis (l), lymphadenitis (3), cyanosis (I), otitis media (1).
for longer
Pi
0.05
than 1 week (1).
M. Kleemola
and
C. Jokinen
r 17
’
18
23
Weeks
I April Marcll May I I Figure 2. Weekly numbers of persons falling ill with acute respiratory infection, and time of the first positive samples during an outbreak of Mycoplasma pneumoniae (MP) infection among the personnel of the Accident and Emergency Department of the Kuopio University Hospital in March-May 1989. Solid line=symptomatic MP-positive persons: their first positive samples (N= 35), dark bar = symptomatic MP-positive persons: onset of respiratory symptoms (the same persons as above; N=35), dotted line=asymptomatic MP-positive persons: their first positive samples (N= S), light bar = symptomatic MP-negative persons: onset of respiratory symptoms (N= 34).
definitely ill. Thirty subjects (31%) eventually received erythromycin and five (5%) doxycycline for 6-l 8 days. Three (3%) received other antibiotics for tonsillitis, wound or urinary tract infection. Twenty-seven persons (28%) were on sick leave for 2-39 d because of acute respiratory infection, fever, or asymptomatic shedding of MP. Only the index patient and one of the two other patients with pneumonia were admitted to hospital.
Duration
of the symptoms and MP shedding
The mean duration in days of the various symptoms was: cough 11.6, rhinitis 14.9, sore throat 9.9, fever 3.8, headache 4.9 and other symptoms 7.0. The duration of these symptoms was very similar irrespective of the MP results or whether antibiotics were given or not. The second sample, taken from the 43 MP-positive persons approximately 23 days after the first, was negative in 40 cases. The three
Mycoplasma
pneumoniae
remaining positives were sampled a third negative. The longest time a subject remained 48 and 57 days. Source
of infection
and transmission
outbreak
219
time, whereupon all were positive for MP was between
in families
The personnel themselves were not aware of any obvious source of infection. The occurrence of symptoms was unrelated to household size: symptomatic employees had an average of 1.6 family contacts and asymptomatic ones, 1.5. There was, however, a difference in the occurrence of clinical illness between the families of symptomatic and asymptomatic individuals. Acute respiratory tract infection or fever during March-May was reported for 53% of the 113 household contacts of symptomatic employees, half of them (30) falling ill before the employees and half (30) later. In the families of asymptomatic persons, only 29% of the 42 family members reported illness (PC 0.01). Some 47% of the family contacts of 43 MP-positive subjects and 46% of those of 54 MP-negative subjects had symptoms, and the average family size was similar in the two groups. Discussion
This study investigated an outbreak of MP infection which occurred among hospital staff. Most of the infections were mild: of 69 symptomatic subjects only three (4%) had pneumonia and 17 (25%) had fever (> 37.5”(Z). In addition, eight (19’/,) of the MP-positive subjects were completely asymptomatic. These figures are in general agreement with published data 4,5,18,19 and further emphasize the mildness of most cases of this disease. If it’had not been for the one severe case in the beginning, the outbreak might well have escaped attention. One aim of the present study was to examine the usefulness of the recently introduced rapid ‘Gen-Probe’ test in monitoring an outbreak of MP infections. Even though a sensitivity as low as 30% with throat swabs has been previously described in our laboratory,13 50% of the symptomatic persons proved to be MP positive. It is thus concluded that, although for individual cases only a positive test is predictive, the ‘Gen-Probe’ test can be used for the detection of an outbreak and for following its course. It is probable that the infections were both community and hospital acquired. A community-acquired origin was suggested by the large number of family contacts whose symptoms started before those of the staff and additional hospital spread indicated by the higher percentage of symptomatic individuals among the staff (70%) compared to family members (45%). Given the low sensitivity of the ‘Gen-Probe’ test using throat swabs, it is possible that within a short time period (2 months) up to 90% of the personnel were infected with MP and 70% had symptoms. Given such an extensive outbreak it is likely that transmission to patients would also have occurred. No evidence of this was obtained; however, such spread may have
220
M. Kleemola
and C, Jokinen
been missed. The patients remained in the hospital for an average of only 5 days, i.e. much less than the incubation period of MP infection” and given the general mildness of the infections, their development following discharge might not have been deemed important. It is also possible that MP infection arising in patients with or convalescing from severe diseases might have resulted in serious manifestations falsely ascribed to the underlying disease alone, How could one prevent the spread of infection among hospital staff and patients? In the light of the present experience this seems a very difficult task. Insisting on sick leave for affected staff might be effective; however, such a course of action is complicated by (a) the many MP positives among asymptomatic individuals, (b) the low sensitivity of the ‘Gen-Probe’ test, and finally (c) the very large proportion of staff who might be infected. For maximum safety the whole ward would need to be closed-but the consequences to patient care could be more deleterious than exposure to MP. Neither the limited sick leave practised in the current episode nor the instigation of antibiotic therapy was successful in stopping the spread of infection” A further difficulty in controlling outbreaks like this is posed by their insidious beginning-when would one decide to institute major control measures? One possibility would be continuous monitoring of all (even mild) respiratory infections in the staff using the ‘Gen-Probe’ test. The first positive finding might then be a signal for more extensive probing of the whole personnel, and for ordering sick leave if several positives are found. This raises several questions: should other respiratory pathogens be monitored in the same way and would this scheme be cost effective? In which wards should it be instituted? These are questions that remain open with present knowledge and call for further investigation. Such studies should also examine the extent of transmission of these infections to patients and their impact on disease and recovery. We thank the personnel of the Accident and Emergency Department of Kuopio University Hospital for their invaluable co-operation, and MS Sari Jokinen as well as other technicians in the laboratory of the National Public Health Institute for their skilful aid in testing the samples. We are especially grateful to Prof. P. Helena Makela for critical evaluation of our manuscript. Our thanks are also due to Mr Jukka Lauronen for the statistical analyses and MS Riitta Brelih for revising our manuscript.
References 1. Denny FW, Clyde WA, Glezen WP. Mycoplasmapneumoniae disease: clinical spectrum, pathophysiology, epidemiology, and control. r Infect Dis 1971; 123: 74-92. 2. Foy HM, Kenny GE, Cooney MK, Allan ID. Long-term epidemiology of infections with Mycoplasma pneumoniae. J Infect Dis 1979; 139: 681-687. 3. Lind K, Bentzon MW. Changes in the epidemiological pattern of Mycoplasma pneumoniae infections in Denmark. A 30 years survey. Epidemiol Infect 1988; 101: 377-386. 4. Smith, Chanock RM, Friedewald WT, Alford RH. Mycoplasma pneumoniae infections in volunteers. Ann NY Acad Sci 1967; 143: 471483.
Mycoplasma
pneumoniae
outbreak
221
5. Broughton R. Infections due to Mycoplasma pneumoniae in childhood. Pediatr Infect Dis J 1986; 5: 71-85. 6. Mogabgab W. Mycoplasma pneumoniae and adenovirus respiratory illnesses in military and university personnel, 1959-1966. Am Rev Respir Dis 1968; 97: 345-358. 7. Steinberg P, White RJ, Fuld SL, Gutekunst RR, Chanock RM, Senterfit LB. Ecology of Mycoplasmapneumoniae infections in marine recruits at Parris Island, South Carolina. Am r Epidemiol 1969; 89: 62273. 8. Fisher B, Yu B, Armstrong D, Magi11 J. Outbreak of Mycoplasma pneumoniae infection among hospital personnel. Am g Med Sci 1978; 276: 205-209. 9. Kashiwagi S, Hayashi J, Nomura H et al. An outbreak of Mycoplasma pneumoniae infections in a hospital in Japan. Kurume Med J 1985; 32: 2933296. 10. Tully JG, Taylor-Robinson D, Cole RM, Rose DL. A newly discovered mycoplasma in the human urogenital tract. Lancet 1981; i: 1288-1291. 11. Baseman JB, Dallo SF, Tully JG, R ose DL. Isolation and characterization of Mycoplasma genital&m strains from the human respiratory tract. J Clin Microbial 1988; 26: 226662269. 12. Harris R, Marmion BP, Varkanis G, Kok T, Lunn B, Martin J. Laboratory diagnosis of Mycoplasmapneumoniae infection. 2. Comparison of methods for the direct detection of specific antigen or nucleic acid sequences in respiratory exudates. Epidemiol Infect 1988;
101: 685-694. 13. Kleemola pneumoniae
SRM, Karjalainen infection: clinical
JE, RCty RKH. Rapid diagnosis of Mycoplasma evaluation of a commercial probe test. 2 Infect Dis 1990;
162: 70-75. 14. Hata D, Kuze F, Mochizuki Y et al. Evaluation of DNA probe test for rapid diagnosis of Mycoplasma pneumoniae infections. J Pediatr 1990; 116: 273-276. 15. Dular R, Kajioka R, Kasatiya S. Comparison of Gen-Probe commercial kit and culture technique for the diagnosis of Mycoplasma pneumoniae infection. r Clin Microbial 1988; 26: 106881069. 16. Tilton RC, Dias F, Kidd H, Ryan RW. DNA probe versus culture for detection of Mycoplasma pneumoniae in clinical specimens. Diagn Microbial Infect Dis 1988; 10: 1099112. 17. Hawkes RA. General principles underlying laboratory diagnosis of viral infections. In: Lennette EH, Schmidt NJ, Eds. Diagnostic Procedures for Viral Rickettsial and Chlamydial Infections. 5th edn. Washington, DC: American Public Health Association 1979: 3542. 18. Biberfeld G, Sterner G. A study of Mycoplasmapneumoniae infections in families. Stand .r Infect Dis 1969; 1: 39-46. 19. Grayston JT, Kenny GE, Foy HM., Kronmal RA, Alexander ER. Epidemiological studies of Mvcoalasma Pneumoniae infections in civilians. Ann NY Acad Sci 1967: 143: 436446. - _ _
