Pediatric Pulmonology 12:158-161 (1992)
Mucoid Pseudomonas aeruginosa Is a Marker of Poor Survival in Cystic Fibrosis Richard L. Henry, MB,
Craig M. Mellis, MB, Lea Petrovic, BA3
BS, FRACP, DIP CLIN EPID,’
BS, FRACP, MPH,’
and
Summary. The aim of this study was to assess the prognostic significance of mucoid and non-mucoid isolates of Pseudomonas aeruginosa (muPs and non-muPs) from the sputa of patientswith cystic fibrosis (CF). Eighty-onechildren with CF who coughed up sputum daily were recruitedand followed over 12 months with frequent sputum cultures. At the end of this observation period they were classified to one of three age-matchedgroups. In 50 mPs was isolated on one or more occasions; 19 grew non-muPs but not muPs, and 12 grew no isolates of Ps aeruginosa. These 81 children and adolescents were followed for a further 8 years or until they died. Twenty-one (42%) of the muPs patients died compared with two (11%) of the non-muPs and one (8%) of the no Ps patients (P< 0.01). Stepwise regression indicated that forced expiratory volume in 1 second (FEV,) had the main predictive effect but that age, Shwachman score and muPs also had a predictive effect. Identificationof mucoidforms of Ps eeruginosa is an unfavorable prognostic factor but the isolation of non-mucoid strains does not appear to be any more important than the isolation of other common respiratory pathogens. Pediatr Pulmonol. 1992; 12:158-161. o 1992Wiley-LissInc.
Key words: Non-mucoid Ps.;no Ps colonization; forced explratory volume in 1 second; Shwachman score; cumulative survival.
INTRODUCTION
The present study reports a survival analysis in the cohort of patients who were classified according to the results of sputum cultures.
Several reports indicate that the lungs of most patients with cystic fibrosis (CF) become colonized with Pseudomonas aeruginosa (Ps aeruginosa) during their lifetime. The clinical significance of this colonization MATERIALS AND METHODS is disputed, with some groups associating colonization by Ps aeruginosa with rapid clinical d e t e r i o r a t i ~ n , ~ ~ “ ~The original cohort’ was selected from 206 children whereas others question the importance of P s aerugi- and adolescents who attended the Cystic Fibrosis Clinic nosa.1*9s10 Part of the problem may be due to a failure to at the Royal Alexandra Hospital for Children between distinguish between mucoid and non-mucoid strains of Ps October, 1979 and September, 1980. The cohort of 82 aeruginosa.” Bums and May” reported that serum pre- children who were identified from this group met the cipitins to Pseudomonus were usually associated with following criteria: each expectorated sputum daily with mucoid strains rather than non-mucoid strains in the spu- cough or during chest percussion and drainage; and each tum and suggested that the mucoid variety is pathogenic. had a Shwachman score performed by the same observer In a previous study Henry et a1.’ assessed the relative on at least one occasion during the 12 month period. As described previously,’ 743 sputum samples were importance of mucoid Ps aeruginosa, non-mucoid P s obtained on an out-patient basis from these 82 patients aeruginosa, and the absence of Ps aeruginosa as indicators of severity of lung disease in age-matched groups of children and adolescents with CF. Both Shwachman From the Departments of Pediatrics’ and Community M e d i ~ i n e Uni,~ score’3 and forced expiratory volume in 1 second (FEV,) versity of Newcastle, Newcastle, and Department of Respiratory Medwere significantly lower in subjects whose sputa yielded icine, The Children’s Hospital, Camperdown,’ Australia. mucoid strains than in those with no Ps aeruginosa or with non-mucoid forms. In that study the identification of Received August 13, 1 9 9 1 ; (revision) accepted for publication October mucoid Ps aeruginosa was an unfavorable prognostic 28, 1991. factor, but isolation of non-mucoid forms of P s aerugi- Address correspondence and reprint requests to Dr. R. L. Henry, nosa did not appear to be any more sinister than the Department of Paediatrics, John Hunter Hospital, Locked Bag 1 , Newcastle Mail Centre 2310. NSW Australia. isolation of other common respiratory pathogens. 0 1992 Wiley-Liss, Inc.
Mucoid Pseudornonas aeruginose in Cystic Fibrosis
159
TABLE 1-Comparison of Baseline Characteristics of Children Whose Sputa Over a One Year Period Showed no Ps aeruginosa, Mucoid Ps aeruginosa, or Non-Mucoid Ps aerualnosa (Without Mucoid Ps aerualnosa) No Ps aenlginom
Number (a)
Sputum cultures Mucoid Ps aeruginosa
Non-mucoid
Ps mginosa
12 (15)
50 (62)
19 (23)
7 (58) 5 (42) 10.5 (8.4- 12.5) 83.8 (80.1-87.4) 79.5 (68.9-90.2)
28 (56) 22 (44) 10.6 (9.6- 11.7) 75.2 (7 1.9-78.5) 64.5 (57.6-7 1.4)
13 (68) 6 (32)
Sex
M (8) F (96) Mean age (years) (95%
CI)
Mean Shwachman score (95%
CI)
Mean FEVl (96 predicted) (95%
CI)
during the 12 month study period. In the microbiology laboratory this sputum was sampled with a sterile swab, taking care to select a purulent portion if present, and placed onto a McConkey and two blood agar plates. One blood agar plate was incubated anaerobically and the remaining plates aerobically overnight at 35°C. The plates were examined the next day and the McConkey plate incubated for a further 24 hours. Mucoid strains of Ps aeruginosa were more readily recognized with prolonged incubation of McConkey plates. We had no data about whether McConkey plates might induce in vitro mucoid production by Ps aeruginosa. The Shwachman score is a system of clinical evaluation of patients with CF and is derived from an appraisal of each of the following four categories: 1) general activities; 2) physical examination; 3) nutrition; and 4) X-ray findings. A total of 100 points represents a perfect score; the lower the score the worse is the state of the patient. Seventy-one of the 82 children and adolescents had a measurement of FEV, with results expressed as percentage predicted. Eight of the remaining 11 patients were unable to perform a forced vital capacity maneuver, while three failed to keep appointments for respiratory function testing. One of the patients (a 16 year old girl whose sputum had grown mucoid Ps aeruginosa) died during the initial 12 month recruitment period. The progress of the remaining 81 patients was observed for the next 8 years, until the end of August, 1988. The outcome factor measured was whether or not each patient was alive or dead. The date of death was recorded. No data were collected on the patients attending the cystic fibrosis clinic who were not enrolled in the study. Statistical analysis was undertaken using the BMDP Statistical Software Manual. l4 The product-limit (KapIan-Meier) method was used in estimation of survival and survival curves. Cox’s proportional hazards regression model was used for the stepwise survival analysis.
10.5 (8.6- 12.4) 85.4 (82.5-88.3) 87.2 (79.4-95.0)
TABLE 2-Outcome of Patients (Dead or Alive) Eight Years After They Had Been Ciassified According to SDutum Cultures No Ps aeruginosa Non-mucoid Ps aeruginosa Mucoid Ps aem&sa
Alive*
Dead
(96)
11 17 29
1 2 21
(8) (11) (42)
* P < 0.01.
RESULTS
We allocated the patients to one of three groups. During the 1 year observation period, between October, 1979 and September, 1980, patients were classified as producing sputum that showed no isolates of Ps aeruginosa (no Ps) , isolates that yielded mucoid Ps aeruginosa on one or more occasions (muPs), and isolates that yielded nonmucoid Ps aeruginosa, but not mucoid strains (nonmuPs). Patients who had isolates of muPs and non-muPs were classified in the mucoid Ps group. Table 1 compares the baseline characteristics of the children in each of the three groups. There were no significant differences between the mean ages of the children in these groups. Based on 95% confidence intervals (95% CI) the mean Shwachman score in the muPs group was significantly lower than in both the non-muPs and the no Ps groups, and the mean FEN, in the muPs group was significantly lower than in the non-muPs group. Table 2 summarizes the mortality rates in the three groups. The poor prognosis for subjects in the muPs group in comparison to both non-muPs and no Ps groups is demonstrated in more detail by cumulative survival curves (Fig. 1). Mucoid Ps aeruginosa was not identified more frequently in older patients (Table 3).-Mucoid Ps aeruginosa was isolated in 57% of 3 to 8 year old children, in 69% of the 9-13 year age group, and in 53% of the 14-19 year age range. The percentages of children in the 3-8, 9-13, and 14-19 year age brackets who died
Henryetai.
160 1A
.#)
.80
‘iii,
.m
1” .a .30
.20
.10
0.0
Fig. 1. Cumulative proportion of patlent survlval related to I% sults of sputum culture. 0, mucold Ps aeruglnma In sputum; m, non-mucoid Ps aeruglnosa;0,no Ps aeruginola.
TABLE 3-Age at Time of Entry to Study Compared With SDutum Isolates heats)
No Psaeruninosa
Mucoid Psaenrainosa
Non-mucoid Psaenu?inosa
3-8 9-13 14-19
4 5 3
16 25 9
6
Age
8
5
over the 8 year observation period were 36%, 28%, and 24% respectively. Stepwise survival analysis on the 70 patients with the covariates age, sex, Shwachman score, FEV,, and sputum showed that lower FEV, and younger age at the time of entry into the study were the two variables associated (P < 0.05) with poorer survival. Since 11 children had not performed spirometry during the recruitment period, we ran a further stepwise analysis without FEV, . Both lower Shwachman score and mucoid Ps aeruginosa were associated with poorer survival (P G 0.05). DISCUSSION
We found a strong relationship between the identification of mucoid Ps aeruginosa in the sputum at some time
during the 1 year recruitment period and mortality over the subsequent 8 years. By contrast, the isolation of nonmucoid isolates of Ps aeruginosa was not a predictor of increased mortality. This may support the belief that mucoid forms of Ps aeruginosa are the pathogenic strains.2,6,I 1.12 Kerem et al.’ reported that “the clinical course and the pulmonary deterioration in cystic fibrosis after Ps aeruginosa colonization is a gradual and variable process.” By contrast, we found a 42% mortality over 8 years in our patients who were colonized with mucoid Ps aeruginosa compared to 8% and 11% in those with no Ps and non-mucoid Ps aeruginosa, respectively. Kerem’s paper does not distinguish between mucoid and non-mucoid strains, so it is impossible to determine whether the isolates of non-mucoid Ps aeruginosa have obscured the deleterious effects of mucoid Ps aeruginosa. Our study does not allow one to make the distinction between causal effect and association. Consequently this data cannot be used to justify treatment of patients who are colonized by mucoid Ps aeruginosa but who do not have clinical or physiological evidence of deterioration. As expected, FEV, was a good predictor of prognosis and Shwachman score also helped predict survival.’4The fact that younger age was a risk factor for poorer prognosis probably reflects the fact that we had a survivor cohort. Adolescents who were 16 years old at the time of entry into the study had already selected themselves out as having a reasonable prognosis whereas those who were 10 years younger included children who would die before reaching the age of 16. The recruitment of patients into our cohort required that they expectorate sputum daily with cough or during chest percussion and drainage. This was done in an attempt to ensure that the airway samples obtained for culture were representative of the lower respiratory tract. Forty percent of patients attending the clinic were enrolled. We had no way of obtaining an accurate figure for how many of the remaining 60% were colonized by Ps aeruginosa. It is interesting to note that the colonization rate by either muPs or non-muPs in our patients was 85%, compared with 82% reported by Kerem et al.’ Our study did not follow isolates on sputum subsequent to the 1 year observation period and did not look at any outcome measure apart from mortality. A different study design would be required to answer whether a change in colonization from no Ps or non-muPs to muPs is associated with deterioration in lung function or clinical status. Our study suggests that the identification of mucoid forms of Ps aeruginosa in the sputa of patients with CF is an unfavorable prognostic factor, but that the identification of non-mucoid strains has no greater significance than if no Pseudomonus is isolated. Most CF clinics have ready access to data about whether or not the airways are colonized with mucoid Ps aeruginosa. In view of this fact and its predictive value for mortality, consideration
Mucoid Pseudomonas eefuginosa in Cystlc Fibrosis
should be given to including the presence or absence of mucoid Ps aeruginosa in future scoring systems assessing the severity of cystic fibrosis. REFERENCES 1. Kerem E,Corey M, Gold R, Levison H. Pulmonary function and
2. 3. 4. 5.
6.
clinical course in patients with cystic fibrosis after pulmonary colonization with Pseudomorns aeruginosa. J Pediatr. 1990; 116:714-719. Henry RL, Dorman DC, Brown J, Mellis C. Mucoid Pseudomon a s aeruginosa in cystic fibrosis. Aust Paediatr J. 1982; 18:4345. Hoiby N. Microbiology of lung infections in cystic fibrosis patients. Acta Paediatr Scand (Suppl.) 1982; 301:33-54. Kulczycki LL, Murphy TM,Bellanti JA. Pseudomorns colonization in cystic fibrosis: A study of 160 patients. JAMA. 1978; 240:30-34. Laraya-Cuasay LR, Cundy KR, Huang NN. Pseudomom carrier rates of patients with cystic fibrosis and members of their families. J Pediatr. 1976; 89:23-26. Hoiby N. Prevalence of mucoid strains of Pseudomom aeruginosa in bacteriological specimens from patients with cystic fibrosis and patients with other diseases. Acta Pathol Microbiol Scand (Suppl.) 1975; 83549-552.
161
7. Hoiby N, Flensborg EW,Beck B, Friis B, Jacobsen V, Jacobsen L. Pseudomom aeruginosa infection in cystic fibrosis. Diagnostic and prognostic significance of Pseudomonas aeruginosa precipitins determined by means of crossed immunoelectrophoreses. Scand J Respir Dis. 1977; 58:65-79. 8. Hoiby N. Pseudomom aeruginosa infection in cystic fibrosis. Acta Pathol Microbiol Scand (Suppl.) 1977; 262:l-94. 9. Beaudry PH, Marks MI, McDougall D, Desmond K, Rangel R. Is anti-Pseudomoms therapy warranted in acute respiratory exacerbations in children with cystic fibrosis? J Pediatr. 1980 97:144147. 10. Gold R, Carpenter S, Heuster H,Corey M,Levison H. Randomized trial of ceftazidime versus placebo in the management of acute respiratory exacerbations in patients with cystic fibrosis. J Pediatr. 1987; 111:907-913. 11. George RH. Pseudomonas infections in cystic fibrosis. Arch Dis Child. 1987; 62:438-439. 12. Burns MW, May JR. Bacterial precipitins in serum of patients with cystic fibrosis. Lancet. 1968; i:270-272. 13. Shwachman H, Kulczcki LL. Long-term study of one hundred and five patients with cystic fibrosis. Am J Dis Child. 1958; 965-15. 14. Brown MB, Engelman L, Hill MA, Jennrich RI.BMDP Statistical Software Manual. Berkley: University of California Press. Vols I and lI (pp 619 and 1233).
Mucoid Pseudomonas aeruginosa Is a Marker of Poor Survival in Cystic Fibrosis Richard L. Henry, MB,
Craig M. Mellis, MB, Lea Petrovic, BA3
BS, FRACP, DIP CLIN EPID,’
BS, FRACP, MPH,’
and
Summary. The aim of this study was to assess the prognostic significance of mucoid and non-mucoid isolates of Pseudomonas aeruginosa (muPs and non-muPs) from the sputa of patientswith cystic fibrosis (CF). Eighty-onechildren with CF who coughed up sputum daily were recruitedand followed over 12 months with frequent sputum cultures. At the end of this observation period they were classified to one of three age-matchedgroups. In 50 mPs was isolated on one or more occasions; 19 grew non-muPs but not muPs, and 12 grew no isolates of Ps aeruginosa. These 81 children and adolescents were followed for a further 8 years or until they died. Twenty-one (42%) of the muPs patients died compared with two (11%) of the non-muPs and one (8%) of the no Ps patients (P< 0.01). Stepwise regression indicated that forced expiratory volume in 1 second (FEV,) had the main predictive effect but that age, Shwachman score and muPs also had a predictive effect. Identificationof mucoidforms of Ps eeruginosa is an unfavorable prognostic factor but the isolation of non-mucoid strains does not appear to be any more important than the isolation of other common respiratory pathogens. Pediatr Pulmonol. 1992; 12:158-161. o 1992Wiley-LissInc.
Key words: Non-mucoid Ps.;no Ps colonization; forced explratory volume in 1 second; Shwachman score; cumulative survival.
INTRODUCTION
The present study reports a survival analysis in the cohort of patients who were classified according to the results of sputum cultures.
Several reports indicate that the lungs of most patients with cystic fibrosis (CF) become colonized with Pseudomonas aeruginosa (Ps aeruginosa) during their lifetime. The clinical significance of this colonization MATERIALS AND METHODS is disputed, with some groups associating colonization by Ps aeruginosa with rapid clinical d e t e r i o r a t i ~ n , ~ ~ “ ~The original cohort’ was selected from 206 children whereas others question the importance of P s aerugi- and adolescents who attended the Cystic Fibrosis Clinic nosa.1*9s10 Part of the problem may be due to a failure to at the Royal Alexandra Hospital for Children between distinguish between mucoid and non-mucoid strains of Ps October, 1979 and September, 1980. The cohort of 82 aeruginosa.” Bums and May” reported that serum pre- children who were identified from this group met the cipitins to Pseudomonus were usually associated with following criteria: each expectorated sputum daily with mucoid strains rather than non-mucoid strains in the spu- cough or during chest percussion and drainage; and each tum and suggested that the mucoid variety is pathogenic. had a Shwachman score performed by the same observer In a previous study Henry et a1.’ assessed the relative on at least one occasion during the 12 month period. As described previously,’ 743 sputum samples were importance of mucoid Ps aeruginosa, non-mucoid P s obtained on an out-patient basis from these 82 patients aeruginosa, and the absence of Ps aeruginosa as indicators of severity of lung disease in age-matched groups of children and adolescents with CF. Both Shwachman From the Departments of Pediatrics’ and Community M e d i ~ i n e Uni,~ score’3 and forced expiratory volume in 1 second (FEV,) versity of Newcastle, Newcastle, and Department of Respiratory Medwere significantly lower in subjects whose sputa yielded icine, The Children’s Hospital, Camperdown,’ Australia. mucoid strains than in those with no Ps aeruginosa or with non-mucoid forms. In that study the identification of Received August 13, 1 9 9 1 ; (revision) accepted for publication October mucoid Ps aeruginosa was an unfavorable prognostic 28, 1991. factor, but isolation of non-mucoid forms of P s aerugi- Address correspondence and reprint requests to Dr. R. L. Henry, nosa did not appear to be any more sinister than the Department of Paediatrics, John Hunter Hospital, Locked Bag 1 , Newcastle Mail Centre 2310. NSW Australia. isolation of other common respiratory pathogens. 0 1992 Wiley-Liss, Inc.
Mucoid Pseudornonas aeruginose in Cystic Fibrosis
159
TABLE 1-Comparison of Baseline Characteristics of Children Whose Sputa Over a One Year Period Showed no Ps aeruginosa, Mucoid Ps aeruginosa, or Non-Mucoid Ps aerualnosa (Without Mucoid Ps aerualnosa) No Ps aenlginom
Number (a)
Sputum cultures Mucoid Ps aeruginosa
Non-mucoid
Ps mginosa
12 (15)
50 (62)
19 (23)
7 (58) 5 (42) 10.5 (8.4- 12.5) 83.8 (80.1-87.4) 79.5 (68.9-90.2)
28 (56) 22 (44) 10.6 (9.6- 11.7) 75.2 (7 1.9-78.5) 64.5 (57.6-7 1.4)
13 (68) 6 (32)
Sex
M (8) F (96) Mean age (years) (95%
CI)
Mean Shwachman score (95%
CI)
Mean FEVl (96 predicted) (95%
CI)
during the 12 month study period. In the microbiology laboratory this sputum was sampled with a sterile swab, taking care to select a purulent portion if present, and placed onto a McConkey and two blood agar plates. One blood agar plate was incubated anaerobically and the remaining plates aerobically overnight at 35°C. The plates were examined the next day and the McConkey plate incubated for a further 24 hours. Mucoid strains of Ps aeruginosa were more readily recognized with prolonged incubation of McConkey plates. We had no data about whether McConkey plates might induce in vitro mucoid production by Ps aeruginosa. The Shwachman score is a system of clinical evaluation of patients with CF and is derived from an appraisal of each of the following four categories: 1) general activities; 2) physical examination; 3) nutrition; and 4) X-ray findings. A total of 100 points represents a perfect score; the lower the score the worse is the state of the patient. Seventy-one of the 82 children and adolescents had a measurement of FEV, with results expressed as percentage predicted. Eight of the remaining 11 patients were unable to perform a forced vital capacity maneuver, while three failed to keep appointments for respiratory function testing. One of the patients (a 16 year old girl whose sputum had grown mucoid Ps aeruginosa) died during the initial 12 month recruitment period. The progress of the remaining 81 patients was observed for the next 8 years, until the end of August, 1988. The outcome factor measured was whether or not each patient was alive or dead. The date of death was recorded. No data were collected on the patients attending the cystic fibrosis clinic who were not enrolled in the study. Statistical analysis was undertaken using the BMDP Statistical Software Manual. l4 The product-limit (KapIan-Meier) method was used in estimation of survival and survival curves. Cox’s proportional hazards regression model was used for the stepwise survival analysis.
10.5 (8.6- 12.4) 85.4 (82.5-88.3) 87.2 (79.4-95.0)
TABLE 2-Outcome of Patients (Dead or Alive) Eight Years After They Had Been Ciassified According to SDutum Cultures No Ps aeruginosa Non-mucoid Ps aeruginosa Mucoid Ps aem&sa
Alive*
Dead
(96)
11 17 29
1 2 21
(8) (11) (42)
* P < 0.01.
RESULTS
We allocated the patients to one of three groups. During the 1 year observation period, between October, 1979 and September, 1980, patients were classified as producing sputum that showed no isolates of Ps aeruginosa (no Ps) , isolates that yielded mucoid Ps aeruginosa on one or more occasions (muPs), and isolates that yielded nonmucoid Ps aeruginosa, but not mucoid strains (nonmuPs). Patients who had isolates of muPs and non-muPs were classified in the mucoid Ps group. Table 1 compares the baseline characteristics of the children in each of the three groups. There were no significant differences between the mean ages of the children in these groups. Based on 95% confidence intervals (95% CI) the mean Shwachman score in the muPs group was significantly lower than in both the non-muPs and the no Ps groups, and the mean FEN, in the muPs group was significantly lower than in the non-muPs group. Table 2 summarizes the mortality rates in the three groups. The poor prognosis for subjects in the muPs group in comparison to both non-muPs and no Ps groups is demonstrated in more detail by cumulative survival curves (Fig. 1). Mucoid Ps aeruginosa was not identified more frequently in older patients (Table 3).-Mucoid Ps aeruginosa was isolated in 57% of 3 to 8 year old children, in 69% of the 9-13 year age group, and in 53% of the 14-19 year age range. The percentages of children in the 3-8, 9-13, and 14-19 year age brackets who died
Henryetai.
160 1A
.#)
.80
‘iii,
.m
1” .a .30
.20
.10
0.0
Fig. 1. Cumulative proportion of patlent survlval related to I% sults of sputum culture. 0, mucold Ps aeruglnma In sputum; m, non-mucoid Ps aeruglnosa;0,no Ps aeruginola.
TABLE 3-Age at Time of Entry to Study Compared With SDutum Isolates heats)
No Psaeruninosa
Mucoid Psaenrainosa
Non-mucoid Psaenu?inosa
3-8 9-13 14-19
4 5 3
16 25 9
6
Age
8
5
over the 8 year observation period were 36%, 28%, and 24% respectively. Stepwise survival analysis on the 70 patients with the covariates age, sex, Shwachman score, FEV,, and sputum showed that lower FEV, and younger age at the time of entry into the study were the two variables associated (P < 0.05) with poorer survival. Since 11 children had not performed spirometry during the recruitment period, we ran a further stepwise analysis without FEV, . Both lower Shwachman score and mucoid Ps aeruginosa were associated with poorer survival (P G 0.05). DISCUSSION
We found a strong relationship between the identification of mucoid Ps aeruginosa in the sputum at some time
during the 1 year recruitment period and mortality over the subsequent 8 years. By contrast, the isolation of nonmucoid isolates of Ps aeruginosa was not a predictor of increased mortality. This may support the belief that mucoid forms of Ps aeruginosa are the pathogenic strains.2,6,I 1.12 Kerem et al.’ reported that “the clinical course and the pulmonary deterioration in cystic fibrosis after Ps aeruginosa colonization is a gradual and variable process.” By contrast, we found a 42% mortality over 8 years in our patients who were colonized with mucoid Ps aeruginosa compared to 8% and 11% in those with no Ps and non-mucoid Ps aeruginosa, respectively. Kerem’s paper does not distinguish between mucoid and non-mucoid strains, so it is impossible to determine whether the isolates of non-mucoid Ps aeruginosa have obscured the deleterious effects of mucoid Ps aeruginosa. Our study does not allow one to make the distinction between causal effect and association. Consequently this data cannot be used to justify treatment of patients who are colonized by mucoid Ps aeruginosa but who do not have clinical or physiological evidence of deterioration. As expected, FEV, was a good predictor of prognosis and Shwachman score also helped predict survival.’4The fact that younger age was a risk factor for poorer prognosis probably reflects the fact that we had a survivor cohort. Adolescents who were 16 years old at the time of entry into the study had already selected themselves out as having a reasonable prognosis whereas those who were 10 years younger included children who would die before reaching the age of 16. The recruitment of patients into our cohort required that they expectorate sputum daily with cough or during chest percussion and drainage. This was done in an attempt to ensure that the airway samples obtained for culture were representative of the lower respiratory tract. Forty percent of patients attending the clinic were enrolled. We had no way of obtaining an accurate figure for how many of the remaining 60% were colonized by Ps aeruginosa. It is interesting to note that the colonization rate by either muPs or non-muPs in our patients was 85%, compared with 82% reported by Kerem et al.’ Our study did not follow isolates on sputum subsequent to the 1 year observation period and did not look at any outcome measure apart from mortality. A different study design would be required to answer whether a change in colonization from no Ps or non-muPs to muPs is associated with deterioration in lung function or clinical status. Our study suggests that the identification of mucoid forms of Ps aeruginosa in the sputa of patients with CF is an unfavorable prognostic factor, but that the identification of non-mucoid strains has no greater significance than if no Pseudomonus is isolated. Most CF clinics have ready access to data about whether or not the airways are colonized with mucoid Ps aeruginosa. In view of this fact and its predictive value for mortality, consideration
Mucoid Pseudomonas eefuginosa in Cystlc Fibrosis
should be given to including the presence or absence of mucoid Ps aeruginosa in future scoring systems assessing the severity of cystic fibrosis. REFERENCES 1. Kerem E,Corey M, Gold R, Levison H. Pulmonary function and
2. 3. 4. 5.
6.
clinical course in patients with cystic fibrosis after pulmonary colonization with Pseudomorns aeruginosa. J Pediatr. 1990; 116:714-719. Henry RL, Dorman DC, Brown J, Mellis C. Mucoid Pseudomon a s aeruginosa in cystic fibrosis. Aust Paediatr J. 1982; 18:4345. Hoiby N. Microbiology of lung infections in cystic fibrosis patients. Acta Paediatr Scand (Suppl.) 1982; 301:33-54. Kulczycki LL, Murphy TM,Bellanti JA. Pseudomorns colonization in cystic fibrosis: A study of 160 patients. JAMA. 1978; 240:30-34. Laraya-Cuasay LR, Cundy KR, Huang NN. Pseudomom carrier rates of patients with cystic fibrosis and members of their families. J Pediatr. 1976; 89:23-26. Hoiby N. Prevalence of mucoid strains of Pseudomom aeruginosa in bacteriological specimens from patients with cystic fibrosis and patients with other diseases. Acta Pathol Microbiol Scand (Suppl.) 1975; 83549-552.
161
7. Hoiby N, Flensborg EW,Beck B, Friis B, Jacobsen V, Jacobsen L. Pseudomom aeruginosa infection in cystic fibrosis. Diagnostic and prognostic significance of Pseudomonas aeruginosa precipitins determined by means of crossed immunoelectrophoreses. Scand J Respir Dis. 1977; 58:65-79. 8. Hoiby N. Pseudomom aeruginosa infection in cystic fibrosis. Acta Pathol Microbiol Scand (Suppl.) 1977; 262:l-94. 9. Beaudry PH, Marks MI, McDougall D, Desmond K, Rangel R. Is anti-Pseudomoms therapy warranted in acute respiratory exacerbations in children with cystic fibrosis? J Pediatr. 1980 97:144147. 10. Gold R, Carpenter S, Heuster H,Corey M,Levison H. Randomized trial of ceftazidime versus placebo in the management of acute respiratory exacerbations in patients with cystic fibrosis. J Pediatr. 1987; 111:907-913. 11. George RH. Pseudomonas infections in cystic fibrosis. Arch Dis Child. 1987; 62:438-439. 12. Burns MW, May JR. Bacterial precipitins in serum of patients with cystic fibrosis. Lancet. 1968; i:270-272. 13. Shwachman H, Kulczcki LL. Long-term study of one hundred and five patients with cystic fibrosis. Am J Dis Child. 1958; 965-15. 14. Brown MB, Engelman L, Hill MA, Jennrich RI.BMDP Statistical Software Manual. Berkley: University of California Press. Vols I and lI (pp 619 and 1233).
