The Transiency of Oropharyngeal Colonization with Gram-Negative Bacilli in Residents of a Skilled Nursing Facility

The Transiency of Oropharyngeal Colonization with Gram-Negative Bacilli in Residents of a Skilled Nursing Facility

The Transiency of Oropharyngeal Colonization with Gram-Negative Bacilli in Residents of a Skilled Nursing Facility· Richard S. Irwin, M.D., F.C.C.P.; ...

2MB Sizes 0 Downloads 5 Views

The Transiency of Oropharyngeal Colonization with Gram-Negative Bacilli in Residents of a Skilled Nursing Facility· Richard S. Irwin, M.D., F.C.C.P.; Susan Whitaker, B.S.; Melvin R. Pratter, M.D., F.C.C.P.; Charles E. Millard, M.D.; Joseph T. Tarpey, M.D.; and R. WiUiam Corwin, M.D.

To determine the duration and penistence of gramnegative bacillary (GNB) oropharyngeal colonization over a specileel period and the risk of subsequent GNB pneumonia developing in nursing home patients, we prospectively cultured for 31 weeks the oropharyDges of patients in a sIdUed nursing facUity. Over a 31-week period, an ave....e of 13.8 percent of the patients showed colonization (weeldy preV81ence rates ranged from 0 to 29 percent). No patient had pneumonia during the study

period. We concluded that the presence of GNB In the oropharynx of our patients is transient, continuaDy cbaDllog over an extended period, and that GNB colonizatioD • a transient occurrence is not direcdy associated with an increased risk of GNB pneumoniL Our data also emphasize the UmitatioDS of previously described singleculture survey studies in predictiog the importaDce of GNB oropharyngeal COlonizatiOIL

The importance of oropharyngeal colonization with aerobic and facultative gram-negative bacilli (GNB) is known only for patients admitted to a medical intensive care unit,' The incidence of GNB colonization rapidly rises in four days from 22 percent of the patients on the first hospital day to 45 percent by the fourth day, and this colonization apparently plays a major role in the subsequent development of nosocomial lower respiratory tract infections. Twenty-three percent of patients with colonization have a lower respiratory tract infection, presumably from GNB, in contrast to 3.3 percent of those patients without colonization. These data suggest that it is not just oropharyngeal colonization with GNB that leaves the patient at risk for subsequent lower respiratory tract infection, but the persistence of that colonization. On the other hand, while the prevalence of oropharyngeal colonization with GNB has been studied in ambulatory alcoholic and diabetic patients.f normal persons," ambulatory, aspiration-prone individuals.t and elderly residents in Institutions.t the importance of colonization in these persons in these settings is unknown. This may be because these studies were limited to single-culture surveys.

A previous study demonstrated that a majority of patients with GNB pneumonias admitted to a large municipal hospital were living in skilled nursing facilities." Consequently, to determine the duration and persistence of GNB colonization and the risk of subsequent GNB pneumonia developing in such nursing home patients, we prospectively cultured, on a weekly basis, the oropharynges of patients in a skilled nursing facility.

·From the Pulmonary Dtvision, Department of Medicine, and Clinical Microbiology Division, Department of Pathology, Rhode Island Hospital, and the Division of Biological and Medical Sciences, Brown University, Providence. This work was supported by a grant from the Rhode Island Foundation and was published in part as an abstract in Clio Res 1977; 25:6fnA. Reprint requests: Dr. Irwin, University of Massachusetts Medical Centef', 55 Lake Avenue North, Worcester 01605

CHEST, 81: 1, JANUARY, 1982

MATERIALS AND METHODS

Subjects for this study were selected from patients in a skilled nursing facility. All were in need of 24-hour nursing and medical supervision. Thirty-two patients, under the care of one of the authors (C.E.M.), were the subjects of this study; none had been in an acute-care facility within the prior three months. There were 20 women and 12 men; their ages ranged from 37 to 97 years, with a mean of 76.9 years. Oropharyngeal cultures were obtained once a week for 31 consecutive weeks in all patients. They were obtained at the same time each day and on the same day each week by the saline gargle method." Each subject gargled 10 ml of isotonic saline for ten seconds and expectorated the solution into a sterile sputum container. Within two hours of collection, gargle solutions were thoroughly mixed with a sterile swab and then inoculated onto blood and MacConkey agar. Because the detailed characterization of normal flora was thought to be too time-consuming and expensive for this study, a semiquantitative method of analysis described by Bartlett et al8 was used. After aerobic incubation in 5 percent CO 2 at 36°C for 48 hours, all aerobic bacteria were estimated from blood agar on a scale of 1 + to 4+ according to the quadrant growth method.s For six consecutive weeks, according to the technique of Rosen-

OROPHARYNGEAL GRAM-NEGATIVE BACILU IN A SKILlED NURSING FACILITY 31

15 patients' (Table 2). GNB colonization was almost uniformly transient No patient showed colonization with the same organism for more than three consecutive weeks; one had colonization for four consecutive weeks but with changing organisms (patient 7, Table 1); and only three patients were colonized 50 percent of the time (patients 4, 6, and 7, Table 1).

thaI, 8 a portion of well-mixed gargle solution was also inoculated into broth media that supports the growth of Enterobacteriaceae, Pseudomonas, and Acinetobacter species to increase the sensitivity in detecting aerobic and facultative GNB. GNB were identifled by standard biochemical techniques (API-20 system, Analytab Products, Inc) ; aD other bacteria were identified by means of standard techniques. Patient infonnation (Tables 1 and 2) on demographic characteristics, physical, mental, and functional status, and diagnoses were obtained from the patient, personal physician, the nursing staff, and the medical record. Data were evaluated statistically by Student's , test and ~ analysis.

Identity of GNB Isolated

REsuLTS GNB Oropharyngeal Colonization During the study period, weekly gargle cultures were obtained in all 32 patients and totalled 992. An average of 13.8 percent of patients showed colonization; weekly prevalence rates ranged from 0 to 29 percent (Fig 1). GNB were never isolated from

The spectrum and frequency of GNB colonizing the oropharynges of patients are listed in Table 3. From the 527 cultures obtained from the 17 patients with colonization, nine different genera were isolated 119 times. Multiples of two or three GNB were isolated 19 percent of the time. Klebsiella was the genus most frequently cultured (29 percent). Vitidans streptococci were present in more than 99 percent of patients with or without colonization.

Table l--Cl&araelerbdu of SNF Paderab ..", CoIoni.do.",. GND· Patient/Age, yr /Sex

Diagnosis

Dentition

Incontinent Ambulation

1/83/F

Heart disease, stroke

Natural

2/94/F

Senility, cataracts

Dentures

3/87/M

Esophageal stricture, Hypothyroidism

Edentulous

+ + +

4/97/F

COPD, Arteriosclerosis, seizure disorder, dislocated shoulder

Natural

+

5/77/F

Chronic renal disease

Natural

6/92/F

Heart disease, stroke, pulmonary embolus

Dentures

+ +

7/94/M

Heart disease, COPD

Dentures

8/72/M

Arteriosclerosis, bilateral amputee, pneumonectomy

Edentulous

9/37/M

Spinal disease

Natural

lO/60/F

Alcoholism, COPD

Dentures

11/86/F

Arteriosclerosis, spinal disease

Dentures

+ +

12/70/F

Parkinsonian, schizophrenia

Dentures

+

+

13/74/F

Arteriosclerosis

Dentures

+

+

14/88/M

Heart disease

Dentures

15/76/F

Stroke, arteriosclerosis, Natural parkinsonian

16/86/F

Stroke

Dentures

17/75/M

Arteriosclerosis, COPD, parkinsonian

Edentulous

Urine

+

Urine/feces

Bedridden

Feces

Wheelchair

Urine/feces

Bed/chair

Decubitus

Antibiotic

+ + +

+

+

+ +

+ +

+ + +

+

+ +

+ Urine

Sedation

+

+ +

+

·SNF - skilled nursing facility.

32 IRWIII ET AL

CHEST,

~1:

1, JANUARY, 1982

Diagnoses

Patient/Age, yr/Sex

Dentition

Incontinent Ambulation Urine/feces

Decubitus

Antibiotic

Sedation

+

+

+ +

l/68/F

Stroke

Natural

2/59/M

Spinal disease, alcoholism

Natural

+

3/87/F

Arteriosclerosis, COPD, heart disease, anemia, arthritis

Dentures

+

4/78/M

Arthritis, arteriosclerosis

Edentulous

+

+

5/7I/F

Arteriosclerosis, Edentulous heart disease, diabetes, parkinsonian

Bed/chair

+

6/85/F

Diverticulitis

Edentulous

+

7/86/M

Arteriosclerosis

Edentulous

8/61/M

Diabetes, stroke, heart disease, COPD

Natural

+ + +

9/77/F

Diabetes, diverticulosis, hiatus hernia

Dentures

1O/77/M

COPD, heart disease, anemia

Edentulous

ll/82/F

Arteriosclerosis, scoliosis

Dentures

12/66/F

Arthritis

Edentulous

13/84/F

Stroke, arthritis

Dentures

14/56/F

Status post hip fracture Natural

IS/77/M

Arteriosclerosis

Bed/chair

Urine/feces

+ +

+ Urine/feces

+

Bed/chair

+

+

+

+

+ +

Dentures

+

+

Bed/chair

+

·SNF-skilled nursing facility.

age, sex, dentition, underlying disease or condition, prior acute illness, inhalational therapy use, or date of previous hospitalization. Anitibiotic use was not a factor associated with colonization. Four patients in both groups took an antibiotic for similar periods during the study. No patient had an upper respi-

Factors Associated with GNB Colonlzation

No disease-specific or function-specific factors (Tables 1 and 2) were found to be associated with GNB oropharyngeal colonization. In comparing patient groups, there was no significant difference in 30

29

r-

28 25

25

23

22

Patients

20

Colonized 15 (%) 10

19 14

13

14 14

1 7

5 ~

~

1 2 3

14 15

14

12

9 7

o

12

11

18

17

15

15

22

4

5

6

7 8

i

i

i i

~

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Weeks

FiCllJlE 1. Oropharyngeal colonization with GNB. Over 3I-week surveillance period, percentage of patients with colonization varied from 0 to 29.

CHEST, 81: 1, JANUARY, 1982

OROPHARYNGEAl GRAll-NEGATIVE BlelW IN A SKILLED NURSING FACIUTY 33

Table S-I'repelle7

0'

GNB ,.""". from.

0r0,,1atIrT-.... C........

Patients I-

No. of Isolates

Organism

%of Total

Klebsiella pneumoniae

35

~29

Pseudomonas species

18

~15

Escherichia ooli

12

::10

Citrobacter freundii

11

~9

Enterobacter cloacae

10

~8

Proteus mirabilis

9

~8

Pseudomonas aeruginosa

8

~7

Enterobacter agglomerans

5

::4

Enterobacter aeroqene«

2

~2

Proteus morganii

2

::2

Serratia lique·aciens

2

~2

Enterobacter hafniae

~1

Acinetobacter lwoffi

f:::l

Pseudomonasmaltophi1lia

~1

Pseudomonasfiuorescen»

~l

Flavobacterium species

~1

Total

119

ratory tract infection during the course of the study. As noted above, viridans streptococci were present in both groups of patients in excess of 99 percent of the gargles obtained. Greater than 90 percent of the time, all subjects had more than 3 + growth of viridans streptococci by the semiquantitative method used in this study. There was no correlation between the quantity of viridans streptococci present in the oropharyngeal Bora and the GNB colonization or duration of colonization. Last, no patient had pneumonia during the course of this study. DISCUSSION

From our results two findings emerge that merit comment. First, the importance of GNB oropharyngeal colonization in our patients was determined. The presence of GNB in gargle cultures was a transient occurrence and not associated with an increased risk of pneumonia during a 3I-week observation period. This transient carriage may explain the following discrepancy in another group of patients': despite the fact that 59 percent of ambulatory alcoholic patients may carry GNB in their oropharynges.l" GNB have been implicated as pathogens in less than 15 percent of patients with community-acquired aspiration pneumonia. 12,11 This

34 IRWlIL ET AL

is surprising, since many alcoholic patients were included in these study populations. Second, our data

emphasize the limitations of single-culture survey studies. The frequency and duration of GNB colonization and the risk of GNB pneumonia developing cannot be answered by those studies. Multipleculture survey studies such as oW'S should be more informative and revealing. Our weekly prevalence data on GNB oropharyngeal colonization demonstrated large variations. These variations were due to transient and random colonization and probably explain diHering point prevalence rates in normal subjects" and other populations described in the literature. Although diHerent sampling and culture methods may also explain the varying prevalence rates in other studies, we believe this to be an unlikely explanation for the weekly variations in our study. We used a reliable and reproducible oropharyngeal culture technique, the saline gargle method, which samples the entire oropharyngeal surface, including the tonsillar crypts and other potentially inaccessible areas. Our results offer no insight to explain why some patients experience transient colonization and others none at all. No disease-specific or functionspecific factors were found to be .associated with GNB oropharyngeal colonization. Moreover, bacterial interference (ie, the ability of bacteria to interfere with the replication of other bacteria) by viridans streptococci did not seem to occur, since their presence did not protect against GNB colonization. These organisms were present in all subjects more than 99 percent of the time. Since naturally acquired viral upper respiratory tract infections have been shown to decrease bacterial adherence to pharyngeal cells in vitro,13 it is possible that differences in intercurrent viral infections in our patients could explain the GNB colonization results. However, no patient had a clinically apparent upper respiratory tract infection. ACKNOWLEDGMENT: The authors thank Head Nurse

Lol1y Calenda, the nursing staff, and other employees at

the Desilets Nursing Home, Warren, RI, for their support and cooperation, and Diana L. Coppolino for her secretarial assistance.

1 Johanson WG Jr, Pierre AK, Sanford JP, Thomas GD. Nosocomial respiratory infections with gram-negative bacilli: the significance of colonization of the respiratory tract. Am Intern Med 1972; 77 :701-06 2 Mackowiak PA, Martin RM, Smith ]W. The role of bacterial interference in the increased prevalence of oropharyngeal gram-negative bacilli among alcoholics and diabetics. Ann Rev Respir Dis 1979; 120:589-93 3 Rosenthal S, Tager lB. Prevalence of gram-negative rods in the normal pharyngeal flora. Ann Intern Moo 1975; CHES~

81: 1, JANUARY, 1982

&'3:355-57

4 Mackowiak PA, Martin RM, Jones SR, Smith IN. Pharyngeal colonization by gram-negative bacilli in aspiration-prone persons. Arch Intern Med 1978; 138: 1224-27 5 Valenti WM, Trudell RG, Bentley DW. Factors predisposing to oropharyngeal colonization with gramnegative bacilli in the aged. N Engl J Med 1978; 298:1108-11 6 Sullivan AG Ir, Dowdle WR, Marine WM, et ale Adult pneumonia in a general hospital: etiology and risk factors. Arch Intern 1972; 129:935-42 7 Johnston DA, Bodey GP. Semiquantitative oropharyngeal culture technique. Appl Microbiol 1970; 20:218-23 8 Bartlett IG, Brewer NS, Ryan KS. Laboratory diagnosis of lower respiratory tract infections. In: Washington JA

Moo

9

10 11 12 13

II (ed). Cumulative techniques and procedures in clinical microbiology, Cumitech 7. American Society for Microbiology, Washington, DC, September 1978:1-15 Syndman DR, Gorbach SL. Sputum: a diagnostic pitfall in alcoholics. Arch Intern Moo 1918; 138:1778-79 Fuxench-L6pez Z, Ramirez-Ronda CH. Pharyngeal flora in ambulatory alcoholic patients: prevalence of gramnegative bacilli. Arch Intern Med 1978; 138:1815-16 Bartlett IG, Gorhach SL, Finegold SM. The bacteriology of aspiration pneumonia. Am J Med 1974; 56:202-07 Lorber B, Swenson RM. Bacteriology of aspiration pneumonia: a prospective study of community- and hospitalacquired cases. Ann Intern Med 1974; 81:329-31 Fainstein V, Musher DM, Catz TR. Bacterial adherence to pharyngeal cells during viral infection. J Infect Dis 1980; 141:172-76

Primary Pulmonary Hypertension The Symposium on Primary Pulmonary Hypertension will be held in Vienna, Austria, March 18-20, organized by the Working Group, "Pulmonary Circulation of the European Society of Cardiology." For information, write to Doz. Dr. J. Mlczoch, Kardiologische UniversitatskIinik, Garnisongasse 13, A1090 Vienna, Austria,

Occupational and Environmental Lung Disease The Division of Pulmonary Disease, Department of Internal Medicine, University of South Florida College of Medicine, will sponsor this course March 20-21. The course wiD take place in the Medical Center Auditorium, Tampa, Florida. Co-sponsor is the Gulf Coast Lung Association of St. Petersburg. For information, contact the Course Director: Keith W. Chandler, M.D., Pulmonary Disease Section (IIIC), 13000 North 30th Street, Tampa, Florida 33612.

CHEST, 81: 1, JANUARY, 1982

OROPHARYNGEAL GRAM-NEGATIVE BACILLI IN A SKILLED NURSING FACIUTY 35