etiologySeven hundred twenty-six patients showing clinical and radiologic findings of community-acquired pneumonia were enrolled in the study; however, 66 patients were subsequently excluded from the study for the following reasons: misdiagnosis at enrollment (n = 59); nosocomial acquisition hospitalization (n = 4); and consent not obtained (n = 3). Thus, 660 patients constituted the final study group. Underlying diseases were found in 324 patients (49%), among which COPD (114 patients), diabetes mellitus (106 patients), and congestive heart disease (58 patients) were the most common.

The subgroup of patients with diabetes mellitus was mainly constituted by patients with type 2 diabetes, which was diagnosed in 100 patients (94%). The mean duration of illness was 8 years, although for 14 patients the diagnosis was established during the present episode of pneumonia. At study entry, the mean plasma glucose level was 238 mg/dL, and the mean hemoglobin A1c value was 8.1%. Diabetes-related complications were common among these patients; thus, 20 patients (19%) had diabetic retinopathy, 18 patients (17%) had diabetic nephropathy, 18 patients (17%) had experienced major mac-rovascular events (eg, ischemic heart disease, stroke, or intermittent claudication), and 5 patients (5%) had peripheral polyneuropathy; for only 56 patients were complications absent. In addition to diet, 59 patients (56%) were receiving treatment with oral agents, and 40 patients (38%) were receiving treatment with insulin (both therapies were simultaneously used in 18 patients); however, during pneumonia 95 patients (90%) needed insulin therapy.

The characteristics of patients with diabetes and community-acquired pneumonia were compared with those of the remaining subgroup of patients, and the results are shown in Table 1. In the univariate analysis, patients with diabetes were significantly older (p = 0.001) and had more severe pneumonia according to the prognostic classification (p < 0.001); in consequence, hospitalization was more frequently required (p 65, the presence of other concomitant comorbid conditions, bacteremia, empyema, and multilobar infiltrates were also found to be independent factors related to mortality. Diabetes was also an independent factor associated with the development of pleural effusion,

diabetesIf we perform a study that is restricted to patients with diabetes (Table 3), we can see that mortality was associated with the presence of concomitant underlying diseases (p = 0.004), some diabetes-related complications, such as nephropathy (p = 0.040) and vasculopathy (p = 0.002), and multilobar infiltrate (p = 0.004). No relation was found with age, sex, length of the disease, period from diagnosis of diabetes, glucose level at entry, hemoglobin A1c level, insulin therapy during pneumonia, bacteremia, pleural effusion, and empyema or complicated effusion. By contrast, in the multivariate analysis (Table 4), only a multilobar infiltrate (p = 0.003) and the simultaneous presence of comorbidities (p = 0.029) were found to be independently associated with mortality. Finally, pleural effusion appeared to be significantly associated with some diabetes-related complications (ie, nephropathy [p = 0.010] and vasculopathy [p = 0.038]), and the presence of other comorbidities in the multivariate analysis.

Microbiological results obtained from patients with diabetes, in comparison with the nondiabetic population, did not show significant differences (Table 5). For both groups, S pneumoniae, which was identified in 225 patients (34%), was the agent most commonly isolated. For the 12 strains of S pneumoniae recovered from patients with diabetes (7 in blood, 1 in pleural fluid, and 4 in sputum), 11 (92%) were susceptible to penicillin (minimum inhibitory concentration, < 0.06 ^g/mL) and 1 had an intermediate susceptibility (minimum inhibitory concentration, 0.12 to 1 ^g/mL). In relation to erythromycin, nine strains (75%) were susceptible, and three were resistant; all 12 strains were sensitive to cefotaxime and quinolones. In comparison, for the 74 isolates of S pneumoniae obtained from patients without diabetes, the rates of resistance were as follows: penicillin, 44% (intermediate, 32%; resistant, 12%); erythromycin, 35%; cefotaxime, 3%; quinolones, 1% (differences not significant). Atypical microorganisms, such as C pneumoniae, M pneumoniae, and L pneumophila, were also frequent etiologic agents. A polymicrobial infection was detected in 10 patients (9%) with diabetes and in 49 patients (9%) without diabetes. For 41 patients with diabetes (29%) and 161 patients without diabetes (39%), the etiology remained unknown.

Table 1—Characteristics of Community-Acquired Pneumonia, and Comparison Between Patients With and Without Diabetes Mellitus

Characteristics Patients With Diabetes (n = 106) Patients Without Diabetes (n = 554) p Value
Age, yr 69 54 0.001
Sex NS
Male 66 359
Female 40 195
Pneumonia severity index
Class I 4(4) 138 (25) < 0.001
Class II 14(13) 154 (28)
Class III 41 (39) 143 (26)
Class IV 35 (33) 95 (17)
Class V 12(11) 24 (4)
Concomitant underlying diseases 56 (53) 219 (40) 0.018
Typical clinical picture 42 (40) 262 (47) NS
WBC count, cells/^L 15100 13430 NS
Bacteremial 10/84(11) 52/464 (10) NS
Multilobar infiltrate 15 (14) 106 (19) NS
Pleural effusion 33 (31) 111 (20) 0.015
Empyema or complicated effusion 9(8) 42 (8) NS
Hospitalization 99 (93) 431 (78) < 0.001
ICU admission 15 (14) 48 (9) NS
Mortality 18 (17) 40 (8) 0.002
Length of hospital stay,! d 10.2 9.1 NS

Table 2—Multivariate Analyses of Factors Associated With Mortality and Development of Pleural Effusion in Patients With Community-Acquired Pneumonia

Variables Odds
Ratio
95% Confidence Interval p Value
Dependent variable-
mortality
Multilobar infiltrate 4.776 2.549-8.949 < 0.001
Concomitant underlying 3.550 1.853-6.804 < 0.001
diseases!
Bacteremiaj 3.403 1.576-7.351 0.002
Age > 65 yr 2.755 1.408-5.392 0.003
Diabetes mellitus 2.137 1.090-4.189 0.027
Empyema or complicated 2.628 1.013-6.818 0.047
effusion
Dependent variable-pleural
effusion
Diabetes mellitus 2.005 1.227-3.277 0.006
Typical clinical picture 1.863 1.271-2.731 0.001

Table 3—Diabetes-Related Findings and Other Characteristics of Patients With Diabetes Mellitus: Univariate Analysis of Factors Associated With Mortality

Variables Patients p Value
Dead (n = 18) Alive
(n = 88)
Age, yr 71 68 NS
Sex NS
Male 10 56
Female 8 32
Concomitant underlying diseases 15 (83) 40 (45) 0.004
Period from diagnosis of diabetes
< 4 yr 10 (44) 52 (59) NS
> 4 yr 8(56) 36 (41)
Glucose level at entry (mg/dl) 267 239 NS
Hemoglobin A1c level (%) 8.3 8.1 NS
Major diabetes-related complications
Retinopathy 4 (22) 16 (18) NS
Nephropathy 6 (33) 11 (13) 0.040
Vasculopathy 8 (44) 10(11) 0.002
Peripheral polyneuropathy 1(6) 4 (5) NS
Insulin therapy during pneumonia 17 (94) 78 (89) NS
Bacteremiaf 3 (19) 7 (9) NS
Multilobar infiltrate 7 (39) 8 (9) 0.004
Pleural effusion 7 (39) 26 (30) NS
Empyema or complicated effusion 3(17) 6(7) NS

Table 4—Multivariate Analyses of Factors Associated With Mortality and Pleural Effusion in Patients With Diabetes Mellitus and Community-Acquired Pneumonia

Variables Odds
Ratio
95% Confidence Interval p Value
Dependent variable-mortality
Multilobar infiltrate 8.847 2.147-36.451 0.003
Concomitant underlying 5.190 diseases! 1.187-22.690 0.029
Dependent variable-pleural effusion Diabetes-related complications
Nephropathy 25.746 2.202-30.1.015 0.010
Vasculopathy 6.626 1.109-39.582 0.038
Concomitant underlying diseases! 3.247 1.103-9.562 0.033

Table 5—Comparison of Causative Agents in Patients With and Without Diabetes Mellitus

Cause Diagnosis of Patients With Diabetes (n = 106) Diagnosis of Patients Without Diabetes (n = 554)
Definite Probable Totalf Definite Probable Totalf
Conventional bacteria 44 (42) 246 (44)
S pneumoniae 30 3 33 (31) 180 12 192 (35)
Haemophilus influenzae 2 2 4 (4) 9 2 11 (2)
Staphylococcus aureus 1 2 3(3) 5 4 9(2)
Other Gram-negative bacilli 2 0 2 (2) 8 10 18(3)
Other Gram-positive cocci 2 0 2 (2) 11 3 14(3)
Anaerobes 0 0 0 (0) 2 0 2 (0.4)
Atypical agents and viruses 32 (30) 196 (35)
C pneumoniae 10 3 13 (12) 42 19 61 (11)
M pneumoniae 3 3 6 (6) 39 13 52 (9)
C burnetii 3 0 3 (3) 7 16 23 (4)
L pneumophila 1 0 1(1) 26 2 28 (5)
C psittaci 0 0 0 (0) 1 3 4(1)
Influenza A virus 2 3 5 (5) 11 7 18(3)
Influenza B virus 2 1 3 (3) 2 2 4(1)
Adenovirus 1 0 1 (1) 1 1 2 (0.4)
Unknown 41 (39) 161 (29)
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s