INT J TUBERC LUNG DIS 18(7):824–830 Q 2014 The Union http://dx.doi.org/10.5588/ijtld.13.0044
Tuberculosis in the intensive care unit: a prospective observational study C. A. Balkema,* E. M. Irusen,* J. J. Taljaard,† C. F. N. Koegelenberg* *Divisions of Pulmonology and †Infectious Diseases, Department of Medicine, University of Stellenbosch & Tygerberg Academic Hospital, Cape Town, Western Cape Province, South Africa SUMMARY S E T T I N G : Data on the determinants of tuberculosis (TB) mortality in the intensive care unit (ICU) are scarce. O B J E C T I V E : To describe factors influencing outcomes of patients admitted with TB requiring mechanical ventilation. D E S I G N : All TB patients admitted to the ICU of an academic hospital in South Africa from January 2012 to May 2013 were enrolled. Disease severity was graded according to the Acute Physiology And Chronic Health Evaluation (APACHE II) score. Comorbid diagnoses, clinical features, radiological and laboratory investigations and outcomes were recorded. R E S U LT S : Of 83 patients (mean age 36.5 6 12.9 years; 45 females; 44 human immunodeficiency virus [HIV] positive) admitted with pulmonary (n ¼ 69) and/or extra-pulmonary (n ¼ 37) TB, 39 died in the ICU
(mortality 44.2%), and a further 10 died during hospitalisation (in-hospital mortality 59.0%). Few clinical parameters, special investigations or other ancillary tests predicted outcome. Only CD4 count ,200 cells/mm3 in HIV-co-infected patients (P ¼ 0.043) and absence of lobar consolidation (P ¼ 0.018) were associated with ICU mortality, whereas a high APACHE II score (22.6 vs. 18.1, P ¼ 0.016) and development of renal failure (P ¼ 0.016) were associated with hospital mortality. C O N C L U S I O N : The mortality of TB patients admitted to the ICU was extremely high. Very few parameters were associated with poor outcome, and no single parameter predicted both ICU and in-patient mortality. K E Y W O R D S : intensive care unit; mortality; tuberculosis
THE GLOBAL INCREASE in the incidence of Mycobacterium tuberculosis, especially in conjunction with the exploding human immunodeficiency virus (HIV) pandemic,1,2 has resulted in an increased number of hospital admissions due to tuberculosis (TB). The incidence of TB in South Africa, especially the Western Cape, is exceptionally high, with 935 new cases per 100 000 population per year.3,4 In South Africa, there are an estimated two million coinfected individuals; about 63% of all newly diagnosed TB cases are human immunodeficiency virus (HIV) co-infected.4,5 Previous studies have shown that approximately 1.5% of adults being treated for active TB in academic hospitals develop respiratory failure requiring admission to an intensive care unit (ICU).6,7 In the early 1990s, TB accounted for 6% of all admissions to the medical ICU of a Cape Town-based academic hospital.8 At this stage, it was also recognised that the rising incidence of TB would place an increased demand on ICUs with limited resources.9,10 The
mortality rate of patients with acute respiratory failure due to pulmonary TB (PTB) is high, estimated at between 60% and 70% (range 26–83) in most studies.8,11–19 This is much higher than that of patients with respiratory failure due to other causes.12 Most previous studies on TB patients admitted to the ICU were not conducted in areas of high TB prevalence, and were retrospective in nature. We aimed to prospectively evaluate patients with active TB admitted to a medical ICU in a large academic hospital in the Western Cape, South Africa, to identify potential prognostic factors and determinants of mortality.
METHODS Study population This prospective observational study was conducted from January 2012 to May 2013 at the Tygerberg Academic Hospital, a 1200-bed facility in Cape Town, South Africa. It is one of two referral centres
Correspondence to: Cecile Balkema, Department of Medicine, University of Stellenbosch, P O Box 19063, Tygerberg, Cape Town 7505, South Africa. Tel: (þ27) 21 938 9423. Fax: (þ27) 21 933 3591. e-mail: [email protected] Article submitted 16 January 2014. Final version accepted 20 February 2014.
825
TB mortality in the ICU
in the city and renders a tertiary service to a population of about 1.5 million people. The study was approved by the Stellenbosch University Health Research Ethics Committee (study number N11/11/ 330), Cape Town, South Africa. Informed consent was provided by all subjects or their legal representative before data collection. All patients with active PTB or extra-pulmonary TB (EPTB) who required mechanical ventilation for at least 24 h were considered candidates for the study. Patients were considered to have active TB if at least two of the following criteria were met: 1) smearpositive for acid-fast bacilli (AFB) or Xpertw MTB/ RIF (Cepheid South Africa, Johannesburg, South Africa) on sputum, tracheal aspirate or any other clinical specimen; 2) culture-positive for M. tuberculosis on sputum, tracheal aspirate or any other clinical specimen; 3) histopathological identification of TB granuloma on biopsied tissues; 4) strong clinical suspicion of active TB; 5) strong radiological evidence for active TB; or 6) pleural fluid with a lymphocyte predominance (.75% lymphocytes and/or lymphocyte to neutrophil ratio .0.75) with adenosine deaminase .30 international units (IU)/l. A strong clinical suspicion of active TB required at least two of four constitutional symptoms (weight loss with accompanying fever, night sweats, productive cough, loss of appetite for .2 weeks) as well as known TB contact or history of previous PTB. We defined strong radiological evidence for active TB as cavitation with or without surrounding infiltrates, involving either the upper lobes or the apical segment of the lower lobes, or ultrasound findings of lymphadenopathy, liver infiltrates or splenic granulomas for which there were no alternative explanations.20,21 Clinical data Apart from documenting patient demographics, we specifically noted the reason for ICU admission, the presence or absence of haemoptysis at presentation, a history of comorbid diseases (including HIV infection, acquired immune deficiency syndrome [AIDS], chronic obstructive lung disease [COPD] and diabetes) and relevant medication use (anti-tuberculosis treatment and antiretroviral therapy [ART]). Laboratory and related investigations Laboratory investigations included white blood cell count, platelet count, serum haemoglobin (Hb), serum albumin, C-reactive protein, creatinine, alanine aminotransferase and serum glucose. Absolute CD4 counts were measured in all HIV-positive patients. The APACHE II score was calculated for all patients during the first 24 h of admission to the ICU.22 Imaging All radiological imaging was documented. Chest
radiographs were reviewed on admission, and patterns were classified as 1) cavitation, 2) miliary, 3) diffuse bronchopneumonia, 4) lobar consolidation, 5) pleural effusion, 6) isolated lymph node enlargement or 7) normal. Management, course and complications All patients were managed according to the standard operating procedures of the unit, and received maximal supportive therapy. The standard combination anti-tuberculosis treatment regimen consisting of isoniazid (INH), rifampicin (RMP), pyrazinamide and ethambutol (EMB) was used unless significant renal or hepatic impairment was present or drug resistance was proven.23 Standard diagnostic criteria for the diagnosis of shock, disseminated intravascular coagulation, renal failure, multi-organ dysfunction syndrome (MODS), acute respiratory distress syndrome (ARDS) and ventilator-associated-pneumonia (VAP) were defined according to accepted international criteria.7,24–27 Patients were categorised as ICU survivors/nonsurvivors and hospital survivors/non-survivors. Statistical analyses Descriptive statistics and v2 (not Yates corrected) or Fisher’s exact tests (where indicated) were performed on dichotomous categorical variables, and t-tests on continuous data. Unless stated otherwise, data are given as means and standard deviation (SD).
RESULTS Eighty-three patients (mean age 36.5 6 12.9 years; 38 males) were admitted to the medical ICU over the study period and were ultimately treated for confirmed TB (three patients were admitted twice, 86 admissions in total). Of these, 44 (53%) were HIVpositive. Of the 83 patients, 69 (83.1%) had active PTB and 37 (44.6%) EPTB (Table 1). Of the 37 EPTB patients, 23 (62.2%) had concomitant active PTB. Thirty-two (38.6%) patients were on anti-tuberculosis treatment at the time of admission; the rest were diagnosed after admission. The mean duration of ICU admission was 11.9 days (range 1–56). The primary reason for ICU Table 1 All cases of active TB (n ¼ 83) Diagnoses
n (%)
Isolated PTB (without overt EPTB)*
46 (55.4)
EPTB* Pleural TB Disseminated TB (multiple sites) or miliary TB TB meningitis TB abdomen TB involving other sites
13 11 7 4 2
* 23 of 37 (62.2%) patients with EPTB had concomitant active PTB. TB ¼ tuberculosis; PTB ¼ pulmonary TB; EPTB ¼ extra-pulmonary TB.
(15.7) (13.3) (8.4) (4.8) (2.4)
826
The International Journal of Tuberculosis and Lung Disease
Table 2
Categorical data: ICU survivors vs. non-survivors (per patient, n ¼ 83)
Parameter Patient demographic and risk Female sex HIV-positive CD4 ,200 cells/mm3, n/N AIDS Diabetes mellitus COPD Anti-tuberculosis treatment on admission TB involvement PTB Isolated PTB EPTB Isolated EPTB Previous PTB Haemoptysis AFB-positive Radiology Cavitation Miliary Diffuse bronchopneumonia Lobar consolidation Pleural effusion Lymph node enlargement alone Normal Complications ARDS Renal failure VAP Septic shock DIC MODS
All (n ¼ 83) n
Survivors (n ¼ 45) n
Non-survivors (n ¼ 38) n
45 44 31/44 27 9 6 32
25 26 15/26 13 5 2 22
20 18 16/18 14 4 4 10
0.89 0.66 5.87 1.46 0.94 2.52 0.37
(0.37–2.11) (0.28–1.57) (1.11–30.95) (0.57–3.61) (0.23–3.79) (0.44–14.64) (0.15–0.95)
1 0.467 0.043* 0.590 1* 0.405* 0.060
69 46 37 14 26 15 32
36 23 22 9 17 9 15
33 23 15 5 9 6 17
1.65 1.47 0.68 0.61 0.51 0.75 1.62
(0.58–6.10) (0.61–3.52) (0.28–1.63) (0.18–1.99) (0.20–1.34) (0.24–2.34) (0.66–3.95)
0.590 0.522 0.262 0.558 0.254 0.841 0.403
12 6 39 10 16 3 5
6 3 17 9 11 2 4
6 3 22 1 5 1 1
1.22 1.20 2.26 0.11 0.47 0.58 0.28
(0.36–4.15) (0.23–6.32) (0.94–5.47) (0.01–0.90) (0.15–1.50) (0.05–6.70) (0.03–2.59)
1 1* 0.108 0.018* 0.307 1* 0.369*
26 31 19 23 15 25
12 15 10 10 8 14
14 16 9 13 7 11
1.66 1.46 1.09 1.82 1.04 0.90
(0.63–4.08) (0.60–3.56) (0.39–3.03) (0.69–4.81) (0.34–3.20) (0.35–2.32)
0.450 0.554 0.920 0.332 0.841 1
OR (95%CI)
P value
* Fisher’s exact test (all other P values calculated using v2 test). ICU ¼ intensive care unit; OR ¼ odds ratio; CI ¼ confidence interval; HIV ¼ human immunodeficiency virus; AIDS ¼ acquired immune-deficiency syndrome; COPD ¼ chronic obstructive pulmonary disease; TB ¼ tuberculosis; PTB ¼ pulmonary TB; EPTB ¼ extra-pulmonary TB; AFB ¼ acid-fast bacilli; ARDS ¼ acute respiratory distress syndrome; VAP ¼ ventilator-acquired pneumonia; DIC ¼ disseminated intravascular coagulation; MODS ¼ multiple organ dysfunction syndrome.
admission was acute respiratory failure in two thirds of patients (n ¼ 56). TB was the sole cause of respiratory failure in 19 patients, most of whom had acute respiratory failure due to massive haemoptysis (n ¼ 14). The clinico-radiological syndrome of community-acquired pneumonia (CAP) was present in 38 (45.8%) patients; half of these presented with septic shock (n ¼ 19). Four of the 56 patients presenting with acute respiratory failure had cardiopulmonary arrest necessitating ICU admission. Other reasons for ICU admission included reduced level of consciousness (n ¼ 7), surgery of the gastrointestinal tract (n ¼ 3) and concomitant disease not directly related to TB (n ¼ 3). ICU mortality was 44.2% (39/86 admissions); only 34/83 patients were alive at the time of hospital discharge (in-hospital mortality 59.0%). The three patients who were readmitted all survived their respective ICU and hospital admissions. Comorbid diseases (COPD, diabetes mellitus or HIV/AIDS) were present in 52 patients (62.7%); only six patients had two comorbidities. Of the 44 patients with HIV co-infection, 18 (40.9%) did not survive: 13/31 patients with a CD4 count ,200 cells/mm3
died (51.6%) vs. 2/13 with a CD4 count 7200 cells/ mm3 (15.4%, P ¼ 0.043). About one third of the patients (34.1%) were on antiretrovirals at the time of admission, and just over half were highly active ART-na¨ıve. Cases with PTB and/or EPTB did not differ significantly with regard to ICU mortality (Table 2). Patients with lobar consolidation were at a lower risk of dying (P ¼ 0.018), whereas a higher APACHE II score tended to have a higher ICU mortality (Table 3). No other radiological or laboratory parameter predicted ICU mortality. There was a significant association between renal failure and in-patient death (Table 4). Moreover, patients who died in hospital had a significantly higher APACHE II score (22.6 6 8.5 vs. 18.1 6 7.4, P ¼ 0.016; Table 5). Thirty-two patients (38.6%) were already started on anti-tuberculosis treatment at the time of admission; the others were diagnosed after admission. The median time until start of treatment for patients who were diagnosed with TB in the ICU was 1.6 days (range 0–17). Tracheal aspirates were obtained from all but three patients; 39.8% were AFB-positive.
TB mortality in the ICU
Table 3
827
Continuous data: ICU survivors vs. non-survivors (per patient, n ¼ 83)
Parameter Age, years APACHE II score CD4, cells/mm3 P:F ratio White cell count, x 109/l Platelet count, x 109/l Haemoglobin, g/dl Serum albumin, g/l C-reactive protein, mg/l Creatinine, lmol/l ALT, U/l s-Glucose, mmol/l
Reference NA NA 600-1500 .300 4.0–11.0 150–400 12.0–15.0* 35–50 ,5 ,90* 5–40 4.4–6.1
All patients (n ¼ 83) mean 6 SD 36.5 20.7 160.2 215.2 13.5 282.8 9.76 25.1 175.2 137.7 133.8 8.80
6 6 6 6 6 6 6 6 6 6 6 6
ICU survivors (n ¼ 45) mean 6 SD
12.9 8.3 163.6 113.2 7.1 165.7 2.41 6.3 94.4 152.0 305.1 4.76
36.8 19.3 198.5 220.0 14.1 307.6 9.52 25.1 168.8 131.6 178.6 8.32
6 6 6 6 6 6 6 6 6 6 6 6
12.6 7.7 180.7 115.5 6.7 175.1 2.55 7.0 93.3 139.4 398.8 4.55
ICU non-survivors (n ¼ 38) mean 6 SD 36.1 22.4 104.8 209.4 12.8 253.4 10.04 25.1 183.0 145.0 84.8 9.35
6 6 6 6 6 6 6 6 6 6 6 6
12.6 8.8 119.1 111.8 7.6 150.9 2.23 5.5 97.0 167.4 138.1 5.00
P value 0.803 0.086 0.060 0.668 0.409 0.138 0.257 0.984 0.531 0.690 0.212 0.330
* For females. ICU ¼ intensive care unit; SD ¼ standard deviation; NA ¼ not available; APACHE II ¼ Acute Physiology and Chronic Health Evaluation II; P:F ratio ¼ arterial partial oxygen pressure to inspired fractional oxygen concentration ratio; ALT ¼alanine aminotransferase.
Mycobacterial cultures obtained from a variety of sites, including tracheal aspirates, were positive in 48 patients (57.8%). Other sources included pleural fluid, cerebrospinal fluid, blood, lymph nodes, ascites, urine and placenta. In culture-negative cases, TB diagnosis was based on a high clinical probability, combined with radiological evidence (n ¼ 14), pleural
fluid analysis (n ¼ 13), cerebrospinal fluid analysis (n ¼ 7) or histology (n ¼ 1). Drug susceptibility testing (DST) for first-line antituberculosis drugs was performed on all positive samples, and DST for RMP was performed on Xpertpositive samples. Two patients had INH monoresistance; one had been previously treated for TB.
Table 4 Categorical data: hospital survivors vs. non-survivors (per patient, n ¼ 83) All (n ¼ 83) n
Survivors (n ¼ 34) n
Non-survivors (n ¼ 49) n
45 44 31/44 27 9 6 32
18 19 11/19 10 2 2 16
27 25 20/25 17 7 4 16
1.09 0.82 2.91 1.28 2.67 1.42 0.55
(0.45–2.62) (0.34–1.98) (0.76–11.09) (0.50–3.28) (0.52–13.71) (0.25–8.24) (0.22–1.34)
1 0.823 0.209 0.791 1* 1* 0.273
69 46 37 14 26 15 32
27 20 15 7 11 7 12
42 26 22 7 15 8 20
1.56 0.79 1.03 0.64 0.92 0.75 1.26
(0.49–4.93) (0.33–1.92) (0.43–2.49) (0.20–2.04) (0.36–2.36) (0.24–2.32) (0.51–3.13)
0.647 0.764 1 0.647 1 0.773 0.653
Radiology Cavitation Miliary Diffuse bronchopneumonia Lobar consolidation Pleural effusion Lymph node enlargement alone Normal
12 6 39 10 16 3 5
5 2 13 6 9 1 3
7 4 26 4 7 2 2
0.97 1.42 1.83 0.41 0.46 0.40 0.44
(0.28–3.35) (0.25–8.24) (0.75–4.45) (0.11–1.60) (0.15–1.40) (0.12–16.13) (0.07–2.79)
1* 1* 0.267 0.304 0.271 1* 0.644*
Complications ARDS Renal failure VAP Septic shock DIC MODS
26 31 19 23 15 25
10 7 6 7 4 8
16 24 13 17 11 17
1.16 3.70 1.69 2.05 2.17 1.73
(0.45–3.01) (1.35–10.09) (0.57–4.99) (0.74–5.67) (0.63–7.51) (0.64–4.63)
0.920 0.016 0.498 0.251 0.340 0.396
Parameter Patient demographic and risk Female sex HIV-positive CD4 ,200 cells/mm3, n/N AIDS Diabetes mellitus COPD Anti-tuberculosis treatment on admission TB involvement PTB Isolated PTB EPTB Isolated EPTB Previous PTB Haemoptysis AFB-positive
OR (95%CI)
P value
* Fisher’s exact test (all other P values calculated using v2 test). ICU ¼ intensive care unit; OR ¼ odds ratio; CI ¼ confidence interval; HIV ¼ human immunodeficiency virus; AIDS ¼ acquired immune-deficiency syndrome; COPD ¼ chronic obstructive pulmonary disease; TB ¼ tuberculosis; PTB ¼ pulmonary TB; EPTB ¼ extra-pulmonary TB; AFB ¼ acid-fast bacilli; ARDS ¼ acute respiratory distress syndrome; VAP ¼ ventilator-acquired pneumonia; DIC ¼ disseminated intravascular coagulation; MODS ¼ multiple organ dysfunction syndrome.
828
The International Journal of Tuberculosis and Lung Disease
Table 5
Continuous data: hospital survivors vs. non-survivors (per patient, n ¼ 83)
Parameter Age, years APACHE II score CD4, cells/mm3 P:F ratio White cell count, x 109/l Platelet count, x 109/l Haemoglobin, g/dl Serum albumin, g/l C-reactive protein, mg/l Creatinine, lmol/l ALT, U/l s-Glucose, mmol/l
Reference NA NA 600-1500 .300 4.0–11.0 150–400 12.0–15.0* 35–50 ,5 ,90* 5–40 4.4–6.1
All patients (n ¼ 83) mean 6 SD 36.5 20.7 160.2 215.2 13.5 282.8 9.76 25.1 175.2 137.7 133.8 8.80
6 6 6 6 6 6 6 6 6 6 6 6
12.9 8.3 163.6 113.2 7.1 165.7 2.41 6.3 94.4 152.0 305.1 4.76
Survivors (n ¼ 34) mean 6 SD 35.2 18.1 175.6 209.3 14.2 321.4 9.94 25.3 162.0 109.6 166.1 8.17
6 6 6 6 6 6 6 6 6 6 6 6
11.6 7.4 135.6 110.0 7.4 177.2 2.39 6.9 86.8 135.2 433.0 4.72
Non-survivors (n ¼ 49) mean 6 SD 37.3 22.6 148.4 219.2 13.1 256.1 9.64 25.0 184.4 157.2 110.4 9.22
6 6 6 6 6 6 6 6 6 6 6 6
13.8 8.5 183.9 116.4 6.9 153.4 2.44 6.0 99.6 162.0 161.9 4.80
P value 0.461 0.016 0.592 0.698 0.480 0.077 0.765 0.866 0.331 0.162 0.456 0.330
* For females. ICU ¼ intensive care unit; SD ¼ standard deviation; NA ¼ not available; APACHE II ¼ Acute Physiology and Chronic Health Evaluation II; P:F ratio ¼ ratio of arterial partial pressure of oxygen to inspired fractional concentration of oxygen; ALT ¼alanine aminotransferase.
Another patient previously treated for TB had resistance to INH and EMB. Most patients (83.1%) were anaemic; half of these had Hb ,9 g/dl. White blood cell (WBC) count was elevated in 56.6% of patients; three had a WBC count ,4 3 109. Hypoalbuminaemia was present in most patients (92.5%). There was a high incidence of renal failure (n ¼ 31, 37.3%), which was associated with an increased risk of dying in hospital. The most common radiological finding was diffuse bronchopneumonia (n ¼ 39, 47.0%), followed by pleural effusion (n ¼ 16, 19.3%). Lobar consolidation (n ¼ 10) was the only radiological parameter statistically significantly associated with ICU survival, but not with in-hospital survival (Table 2). Neither ARDS (n ¼ 26, 35.6%) nor MODS (n ¼ 25, 30.1%) was associated with increased mortality. The ICU course was complicated by the development of VAP in 19 patients (27.7%), which did not adversely influence outcome.
DISCUSSION In this prospective observational study, we found high ICU and in-hospital mortality among patients admitted to the ICU with TB, at around 44% and 59%, respectively. Few clinical parameters, special investigations or other ancillary tests predicted outcome. Only CD4 ,200 cells/mm3 in HIV-co-infected patients (P ¼ 0.043) and the absence of lobar consolidation (P ¼ 0.018) were associated with ICU non-survival, whereas a high APACHE II score (P ¼ 0.016) and the development of renal failure (P ¼ 0.016) were associated with hospital non-survival. Our in-hospital mortality rates of acute respiratory failure secondary to TB requiring mechanical ventilation are in line with other published data, and are much higher than those of patients with respiratory failure due to non-tuberculous pneumonia.8,11–19 They in fact approximate the mortality rate for ARDS,
and are much higher than predicted by the APACHE II score. Based on our mean APACHE II score of 20.7, expected mortality would be 30–40%.22 Although the APACHE II score may underestimate mortality in this patient group,28 we did observe a significant difference in score between survivors and non-survivors. These findings are consistent with previously published retrospective data.8 The ICU mortality observed for patients with PTB and/or EPTB did not differ significantly between groups, nor did smear positivity or the diagnosis of miliary TB influence outcome. Due to the small numbers, we cannot comment on the effect of drug resistance on outcome. Several studies have suggested a link between delays in initiating anti-tuberculosis treatment and mortality.12,14,29 Due to the high TB incidence in our setting, we always had a high index of suspicion, particularly in HIV-positive patients, where presentation is often atypical, with a low threshold for initiating empiric anti-tuberculosis treatment. While 38.6% of patients were already on anti-tuberculosis treatment on admission to the ICU, the median time to start of treatment was only 1.6 days. Acute renal failure is a well described complication of critical illness; it is common in the ICU and associated with high ICU mortality. More than a third of our patients had acute renal failure, some requiring dialysis, which was associated with increased risk of dying; this is in keeping with previously published ICU data.12,15,18 Other complications, such as septic shock, ARDS, MODS and VAP, although common, were not linked to poor outcome, as described previously.11,12,14,18 The white blood cell count in our patient population was elevated in 56.6% of patients, and was much higher than in a Brazilian study conducted in a similar setting.18 This can potentially be explained by the high incidence of CAP syndrome, found in 45.8% of admissions; however, no association with mortality
TB mortality in the ICU
was observed. We also did not find an association between low platelet count, low serum albumin, low Hb level, raised liver enzymes or lymphopaenia and poor outcomes.13,14,29–31 Although early retrospective data showed a detrimental effect of HIV on outcome,32 we found no association between the diagnosis of HIV or AIDS and mortality, in keeping with recent retrospective data from Brazil.18 ICU mortality among the HIVinfected subgroup was 40.9%, which is lower than the overall ICU mortality of 44.2%. A marked difference in mortality was observed when stratifying patients according to CD4 count: 13/31 patients with CD4 count ,200 cells/mm3 died (51.6%) vs. 2/13 with CD4 count 7200 cells/mm3 (15.4%, P ¼ 0.043). Profound decreases in CD4 counts have been reported even in the absence of HIV infection, and caution should be used when interpreting data on CD4 counts in general, and particularly in the acutely ill patient.33 In an ICU study conducted by Feeney et al., of 102 ICU patients, only three of whom were HIV-positive, 41% had CD4 levels ,400 cells/mm3 and 29% had ,300 cells/ mm3.34 Low CD4 cell counts in patients diagnosed with TB and subsequent elevation in CD4 count after initiation of anti-tuberculosis treatment have been described in several case series, suggesting that TB itself has an impact on CD4 cell homeostasis.35,36 Contrary to previously published data, lack of cavitation was not linked to poor outcome.31 On the contrary, consolidation was associated with improved ICU, but not hospital outcome. The less extensive lung involvement in this group, with better arterial partial oxygen pressure to inspired fractional oxygen concentration ratios (P:F ratios), might explain the better outcome, as the most common presentation was diffuse bronchopneumonic infiltrates, indicating diffuse and potentially non-reversible lung damage. TB is known to be a slowly responding disease, but the more localised lung involvement may have resulted in better outcomes. Although retrospective case studies have been published from both low- and high-income countries, there is an absence of prospective data on TB in the ICU. Ours is the first prospective study to be conducted in an area of very high TB prevalence. We were able to enrol a large number of patients over a relatively short time period, ensuring that standard of care remained unchanged during the study period and thus limiting the effect of changes in critical care over the years. The high incidence of TB-HIV coinfection made it possible to compare both HIVinfected and non-HIV-infected patients with TB. The design also enabled us to look at a large number of variables, including incidence of multidrug-resistant TB and both PTB and EPTB, whereas previous retrospective data were either unavailable or incomplete with regard to these variables. All of our
829
patients required mechanical ventilation, whereas in previous studies the number of ventilated patients was variable. Limited ICU resources may have introduced a selection bias toward healthier patients being admitted to the ICU. The tendency to admit healthier patients with a better prognosis to the ICU could account for the lack of significance of certain parameters in predicting mortality. Delays in referral — both before admission to hospital and between the emergency department and the ICU — may also have contributed to a delay in diagnosis and treatment of TB and to the high incidence of acute renal failure and other organ failures. Another potential limitation is that 38.6% of patients were already started on antituberculosis treatment at the time of admission. In conclusion, we found that both ICU and inhospital mortality of TB patients admitted to the ICU was extremely high. Few clinical parameters, special investigations or other ancillary tests predicted outcome. Only CD4 ,200 cells/mm3 in HIV-coinfected patients and absence of lobar consolidation were associated with mortality in the ICU, whereas a high APACHE II score and the development of renal failure were associated with hospital mortality. Conflict of interest: none declared.
References 1 Bock N, Reichman L B. Tuberculosis and HIV/AIDS: epidemiological and clinical aspects (World Perspective). Semin Respir Crit Care Med 2004; 25: 337–345. 2 World Health Organization. Global tuberculosis control. Report 2013. WHO/HTM/TB/2013.11. Geneva, Switzerland: WHO, 2013. http://www.who.int/tb/publications. Accessed March 2014. 3 Western Cape Department of Health HIV/AIDS/STI and TB directorate. Action plan for the enhanced response to improve TB control in the Western Cape. Report 2012. Cape Town, South Africa: DoH, 2012. 4 Corbett E L, Watt C J, Walker N, et al. The growing burden of tuberculosis: global trends and interactions with the HIV pandemic. Arch Int Med 2003; 163: 1009–1021. 5 Joint United Nations Programme on HIV/AIDS. Report on the global AIDS epidemic 2013. UNAIDS/JC2502/1/E. Geneva, Switzerland: UNAIDS, 2013. http://www.unaids.org/en/media/ unaids/contentassets/documents/epidemiology/2013/gr2013/ UNAIDS_Global_Report_2013_en.pdf. Accessed March 2014. 6 Levy H, Kallenbach J M, Feldman C, Thorburn J R, Abramowitz J A. Acute respiratory failure in active tuberculosis. Crit Care Med 1987; 15: 221–225. 7 Bernard G, Artigas A, Carlet, J, et al. The American-European consensus conference on ARDS: definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994; 149: 818. 8 Hayhurst M D, Kussman B D, Potgieter P D. Acute respiratory failure due to pulmonary tuberculosis: a five year analysis in an intensive care unit. S Afr J Crit Care 1994; 10: 4–7. 9 Potgieter P D. The influence of the tuberculosis epidemic on ICU resources. S Afr Med J 1995; 85: 1196. 10 Bateman E D. ICU care for tuberculosis patients—fielding on the boundary. S Afr Med J 1997; 87: 635–637. 11 Lee P L, Jerng J S, Chang Y L, et al. Patient mortality of active
830
12
13
14
15
16
17
18
19
20
21
22
23
24
The International Journal of Tuberculosis and Lung Disease
pulmonary tuberculosis requiring mechanical ventilation. Eur Respir J 2003; 22: 141–147. Penner C, Roberts D, Kunimoto D, Manfreda J, Long R. Tuberculosis as a primary cause of respiratory failure requiring mechanical ventilation. Am J Respir Crit Care Med 1995; 151: 867–872. Frame R N, Johnson M C, Eichenhorn M S, Bower G C, Popovich J. Active tuberculosis in the medical intensive care unit: a 15-year retrospective analysis. Crit Care Med 1987; 15: 1012. Zahar J-R, Azoulay E, Klement E, et al. Delayed treatment contributes to mortality in ICU patients with severe active pulmonary tuberculosis and acute respiratory failure. Intensive Care Med 2001; 27: 513–520. Erbes R, Oettel K, Raffenberg M, Mauch H, Schmidt-Ioanas M, Lode H. Characteristics and outcome of patients with active pulmonary tuberculosis requiring intensive care. Eur Respir J 2006; 27: 1223–1228. Ryu Y J, Koh W J, Kang E H, et al. Prognostic factors in pulmonary tuberculosis requiring mechanical ventilation for acute respiratory failure. Respirology 2007; 12: 406–411. Lin S M, Wang T Y, Liu W T, et al. Predictive factors for mortality among non-HIV infected patients with pulmonary tuberculosis and respiratory failure. Int J Tuberc Lung Dis 2009; 13: 335–340. Silva D R, Menegotto D M, Schulz L F, Gazzana M B, Dalcin P T. Mortality among patients with tuberculosis requiring intensive care: a retrospective cohort study. BMC Infect Dis 2010, 10: 54–61. Alshimemeri A A, Arabi Y M, Al-Jahdali H, Olayan A, Al Harbi O, Memish Z. Clinical presentation and outcome of patients diagnosed with active pulmonary tuberculosis in a large critical care unit. Crit Care Shock 2011; 14: 1–6. Geng E, Kreiswirth B, Burzynski J, Schluger N W. Clinical and radiographic correlates of primary and reactivation tuberculosis: a molecular epidemiology study. JAMA 2005; 293: 2740– 2745. Andreu J, Caceres J, Pallisa E, Martinez-Rodriguez. Radiological manifestations of pulmonary tuberculosis. Eur J Radiol 2004; 51: 139–149. Knaus W A, Draper E A, Wagner D P, Zimmerman J E. APACHE II: a severity of disease classification system. Crit Care Med 1985; 13: 818–829. Koegelenberg C F, Nortje A, Lalla U, et al. The pharmacokinetics of enteral antituberculosis drugs in patients requiring intensive care. S Afr Med J 2013; 103: 394–398. Bone R C, Balk R A, Cerra F B, et al. Definitions for sepsis and
25
26
27
28
29
30
31
32
33
34
35
36
organ failure and guidelines for the use of innovative therapies in sepsis. Chest 1992; 101: 1644–1655. Bellomo R, Ronco C, Kellum J A, Mehta R L, Palevsky P. Acute renal failure – definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 2004; 8: R204– R212. ARDS Definition Task Force; Ranieri V M, Rubenfeld G D, Thompson B T, et al. Acute respiratory distress syndrome: the Berlin Definition. JAMA 2012; 307: 2526–2533. American Thoracic Society. Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Respir Crit Care Med 2005; 171: 388–416. Roodt A, Smith C, Feldman C, Levy H, Kallenbach J M. APACHE II severity of illness score in patients with severe active pulmonary tuberculosis. S Afr J Crit Care 1990; 6: 13– 14. Sacks V, Pendle S. Factors related to in-hospital deaths in patients with tuberculosis. Arch Intern Med 1998; 158: 1916– 1922. Maartens G, Willcox P A, Benatar S R. Miliary Tuberculosis: rapid diagnosis, hematologic abnormalities, and outcome in 109 treated adults. Am J Med 1990; 89: 291–296. Barnes P F, Leedom J M, Chan L S, et al. Predictors of shortterm prognosis in patients with pulmonary tuberculosis. J Infect Dis 1988; 158: 366–371. Stoneburner R, Laroche E, Prevots R, et al. Survival in a cohort of human immunodeficiency virus-infected tuberculosis patients in New York City: implications for the expansion of the AIDS case definition. Arch Intern Med 1992; 152: 2033–2037. Aldrich J, Gross R, Adler M, King K, MacGregor R R, Gluckman S J. The effect of acute severe illness on CD 4þ lymphocyte counts in non-immunocompromised patients. Arch Intern Med 2000; 160: 715–716. Feeney C, Bryzman S, Kong L, Brazil H, Deutsch R, Fritz L C. T-lymphocyte subsets in acute illness. Crit Care Med 1995; 23: 1680–1685. Skogmar S, Schon T, Balcha T T, et al. CD4 cell levels during treatment for tuberculosis (TB) in Ethiopian adults and clinical markers associated with CD4 lymphocytopenia. PLOS ONE 2013; 8: e83270. Andersen A B, Range N S, Changalucha J, et al. CD4 lymphocyte dynamics in Tanzanian pulmonary tuberculosis patients with and without HIV co-infection. BMC Infect Dis 2012; 12: 66.
TB mortality in the ICU
i
RESUME
Il existe peu de donne´ es sur les d´eterminants de la mortalit´e de la tuberculose (TB) dans les services de soins intensifs (ICU). O B J E C T I F : D´ecrire les caract´eristiques influen¸cant le devenir des patients admis en ICU pour une TB n´ecessitant une ventilation assist´ee. S C H E´ M A : Tous les patients admis a` l’ICU d’un hopital ˆ universitaire d’Afrique du Sud avec une TB entre janvier 2012 et mai 2013 ont e´ t´e inclus. La gravit´e de la TB a e´ t´e e´ valu´ee selon le score APACHE II. Nous avons enregistr´e les diagnostics de comorbidit´e, les signes cliniques et radiologiques ainsi que les r´esultats du laboratoire. R E´ S U LT A T S : De 83 patients (age ˆ 36,5 6 12,9 ans, 45 femmes, 44 positifs pour le virus de l’immunod´eficience humaine [VIH]) admis avec une TB pulmonaire (n ¼ 69) et/ou extra-pulmonaire (n ¼ 37), 39 patients sont CONTEXTE :
d´ec´ed´es en ICU (mortalit´e 44,2%) ; 10 autres patients sont d e´ ce´ d e´ s pendant l’hospitalisation (mortalit e´ hospitali`ere globale 59%). Peu de param`etres cliniques ou d’autres examens permettaient de pr´evoir le r´esultat. Seul le comptage des CD4 ,200 cellules/mm3 chez les patients coinfect´es par le VIH (P ¼ 0,043) et l’absence de consolidation lobaire (P ¼ 0,018) e´ taient associ´es a` une mortalit´e en ICU, tandis qu’un score APACHE II e´ lev´e (22,6 contre 18,1 ; P ¼ 0,016) et la survenue d’une insuffisance r´enale aigu¨e (P ¼ 0,016) e´ taient associ´es a` une mortalit´e hospitali`ere. C O N C L U S I O N : La mortalit´e des patients tuberculeux admis a` l’ICU est extrˆemement e´ lev´ee. Tr`es peu de param`etres sont associ´es a` une e´ volution d´efavorable, et aucun param`etre n’a pu pr´esager a` lui seul de la mortalit´e en ICU et en service hospitalier. RESUMEN
Existen pocos datos sobre los factores determinantes de la mortalidad por tuberculosis (TB) en las unidades de cuidado intensivo (ICU). O B J E T I V O: Describir las caracter´ısticas que influyen sobre el desenlace cl´ınico de los pacientes tuberculosos hospitalizados que precisan ventilacion ´ meca´nica. M E´ T O D O: Se incluyeron en el estudio todos los pacientes tuberculosos que ingresaron a la ICU de un hospital universitario en Sura´frica entre enero del 2012 y mayo del 2013. Se evaluo´ la gravedad de la enfermedad mediante la puntuacion ´ APACHE II. Se consignaron los diagnosticos ´ de enfermedades concurrentes, las caracter´ısticas cl´ınicas y radiogra´ficas, las investigaciones de laboratorio y los desenlaces cl´ınicos. R E S U LT A D O S: Ingresaron a la ICU 83 pacientes con TB pulmonar (n ¼ 69) y extrapulmonar o ambas (n ¼ 37) 45 mujeres; 44 pacientes (edad 36,5 6 12,9 anos; ˜ positivos al virus de la inmunodeficiencia humana [VIH]). Treinta y nueve pacientes fallecieron en la M A R C O D E R E F E R E N C I A:
unidad (mortalidad 44,2%); otros 10 pacientes fallecieron durante la hospitalizacion ´ (mortalidad intrahospitalaria 59,0%). Pocos para´ metros, investigaciones especiales u otras pruebas complementarias pronosticaron el desenlace. Solo un recuento de c´elulas CD4 ,200 c´elulas/mm3 en los pacientes coinfectados por el VIH (P ¼ 0,043) y la ausencia de consolidacion ´ lobular (P ¼ 0,018) se asociaron con mortalidad en la unidad; sin embargo, una alta puntuacion ´ APACHE II (22,6 contra 18,1; P ¼ 0,016) y la aparicion ´ de insuficiencia renal (P ¼ 0,016) se asociaron con mortalidad intrahospitalaria. ´ N: La mortalidad de los pacientes con CONCLUSIO diagnostico ´ de TB que ingresan a la ICU es extremadamente alta. Muy pocos criterios se asociaron con este desenlace desfavorable y ninguno de ellos pronostico´ la mortalidad general, en la unidad y durante la hospitalizacion. ´
Tuberculosis in the intensive care unit: a prospective observational study C. A. Balkema,* E. M. Irusen,* J. J. Taljaard,† C. F. N. Koegelenberg* *Divisions of Pulmonology and †Infectious Diseases, Department of Medicine, University of Stellenbosch & Tygerberg Academic Hospital, Cape Town, Western Cape Province, South Africa SUMMARY S E T T I N G : Data on the determinants of tuberculosis (TB) mortality in the intensive care unit (ICU) are scarce. O B J E C T I V E : To describe factors influencing outcomes of patients admitted with TB requiring mechanical ventilation. D E S I G N : All TB patients admitted to the ICU of an academic hospital in South Africa from January 2012 to May 2013 were enrolled. Disease severity was graded according to the Acute Physiology And Chronic Health Evaluation (APACHE II) score. Comorbid diagnoses, clinical features, radiological and laboratory investigations and outcomes were recorded. R E S U LT S : Of 83 patients (mean age 36.5 6 12.9 years; 45 females; 44 human immunodeficiency virus [HIV] positive) admitted with pulmonary (n ¼ 69) and/or extra-pulmonary (n ¼ 37) TB, 39 died in the ICU
(mortality 44.2%), and a further 10 died during hospitalisation (in-hospital mortality 59.0%). Few clinical parameters, special investigations or other ancillary tests predicted outcome. Only CD4 count ,200 cells/mm3 in HIV-co-infected patients (P ¼ 0.043) and absence of lobar consolidation (P ¼ 0.018) were associated with ICU mortality, whereas a high APACHE II score (22.6 vs. 18.1, P ¼ 0.016) and development of renal failure (P ¼ 0.016) were associated with hospital mortality. C O N C L U S I O N : The mortality of TB patients admitted to the ICU was extremely high. Very few parameters were associated with poor outcome, and no single parameter predicted both ICU and in-patient mortality. K E Y W O R D S : intensive care unit; mortality; tuberculosis
THE GLOBAL INCREASE in the incidence of Mycobacterium tuberculosis, especially in conjunction with the exploding human immunodeficiency virus (HIV) pandemic,1,2 has resulted in an increased number of hospital admissions due to tuberculosis (TB). The incidence of TB in South Africa, especially the Western Cape, is exceptionally high, with 935 new cases per 100 000 population per year.3,4 In South Africa, there are an estimated two million coinfected individuals; about 63% of all newly diagnosed TB cases are human immunodeficiency virus (HIV) co-infected.4,5 Previous studies have shown that approximately 1.5% of adults being treated for active TB in academic hospitals develop respiratory failure requiring admission to an intensive care unit (ICU).6,7 In the early 1990s, TB accounted for 6% of all admissions to the medical ICU of a Cape Town-based academic hospital.8 At this stage, it was also recognised that the rising incidence of TB would place an increased demand on ICUs with limited resources.9,10 The
mortality rate of patients with acute respiratory failure due to pulmonary TB (PTB) is high, estimated at between 60% and 70% (range 26–83) in most studies.8,11–19 This is much higher than that of patients with respiratory failure due to other causes.12 Most previous studies on TB patients admitted to the ICU were not conducted in areas of high TB prevalence, and were retrospective in nature. We aimed to prospectively evaluate patients with active TB admitted to a medical ICU in a large academic hospital in the Western Cape, South Africa, to identify potential prognostic factors and determinants of mortality.
METHODS Study population This prospective observational study was conducted from January 2012 to May 2013 at the Tygerberg Academic Hospital, a 1200-bed facility in Cape Town, South Africa. It is one of two referral centres
Correspondence to: Cecile Balkema, Department of Medicine, University of Stellenbosch, P O Box 19063, Tygerberg, Cape Town 7505, South Africa. Tel: (þ27) 21 938 9423. Fax: (þ27) 21 933 3591. e-mail: [email protected] Article submitted 16 January 2014. Final version accepted 20 February 2014.
825
TB mortality in the ICU
in the city and renders a tertiary service to a population of about 1.5 million people. The study was approved by the Stellenbosch University Health Research Ethics Committee (study number N11/11/ 330), Cape Town, South Africa. Informed consent was provided by all subjects or their legal representative before data collection. All patients with active PTB or extra-pulmonary TB (EPTB) who required mechanical ventilation for at least 24 h were considered candidates for the study. Patients were considered to have active TB if at least two of the following criteria were met: 1) smearpositive for acid-fast bacilli (AFB) or Xpertw MTB/ RIF (Cepheid South Africa, Johannesburg, South Africa) on sputum, tracheal aspirate or any other clinical specimen; 2) culture-positive for M. tuberculosis on sputum, tracheal aspirate or any other clinical specimen; 3) histopathological identification of TB granuloma on biopsied tissues; 4) strong clinical suspicion of active TB; 5) strong radiological evidence for active TB; or 6) pleural fluid with a lymphocyte predominance (.75% lymphocytes and/or lymphocyte to neutrophil ratio .0.75) with adenosine deaminase .30 international units (IU)/l. A strong clinical suspicion of active TB required at least two of four constitutional symptoms (weight loss with accompanying fever, night sweats, productive cough, loss of appetite for .2 weeks) as well as known TB contact or history of previous PTB. We defined strong radiological evidence for active TB as cavitation with or without surrounding infiltrates, involving either the upper lobes or the apical segment of the lower lobes, or ultrasound findings of lymphadenopathy, liver infiltrates or splenic granulomas for which there were no alternative explanations.20,21 Clinical data Apart from documenting patient demographics, we specifically noted the reason for ICU admission, the presence or absence of haemoptysis at presentation, a history of comorbid diseases (including HIV infection, acquired immune deficiency syndrome [AIDS], chronic obstructive lung disease [COPD] and diabetes) and relevant medication use (anti-tuberculosis treatment and antiretroviral therapy [ART]). Laboratory and related investigations Laboratory investigations included white blood cell count, platelet count, serum haemoglobin (Hb), serum albumin, C-reactive protein, creatinine, alanine aminotransferase and serum glucose. Absolute CD4 counts were measured in all HIV-positive patients. The APACHE II score was calculated for all patients during the first 24 h of admission to the ICU.22 Imaging All radiological imaging was documented. Chest
radiographs were reviewed on admission, and patterns were classified as 1) cavitation, 2) miliary, 3) diffuse bronchopneumonia, 4) lobar consolidation, 5) pleural effusion, 6) isolated lymph node enlargement or 7) normal. Management, course and complications All patients were managed according to the standard operating procedures of the unit, and received maximal supportive therapy. The standard combination anti-tuberculosis treatment regimen consisting of isoniazid (INH), rifampicin (RMP), pyrazinamide and ethambutol (EMB) was used unless significant renal or hepatic impairment was present or drug resistance was proven.23 Standard diagnostic criteria for the diagnosis of shock, disseminated intravascular coagulation, renal failure, multi-organ dysfunction syndrome (MODS), acute respiratory distress syndrome (ARDS) and ventilator-associated-pneumonia (VAP) were defined according to accepted international criteria.7,24–27 Patients were categorised as ICU survivors/nonsurvivors and hospital survivors/non-survivors. Statistical analyses Descriptive statistics and v2 (not Yates corrected) or Fisher’s exact tests (where indicated) were performed on dichotomous categorical variables, and t-tests on continuous data. Unless stated otherwise, data are given as means and standard deviation (SD).
RESULTS Eighty-three patients (mean age 36.5 6 12.9 years; 38 males) were admitted to the medical ICU over the study period and were ultimately treated for confirmed TB (three patients were admitted twice, 86 admissions in total). Of these, 44 (53%) were HIVpositive. Of the 83 patients, 69 (83.1%) had active PTB and 37 (44.6%) EPTB (Table 1). Of the 37 EPTB patients, 23 (62.2%) had concomitant active PTB. Thirty-two (38.6%) patients were on anti-tuberculosis treatment at the time of admission; the rest were diagnosed after admission. The mean duration of ICU admission was 11.9 days (range 1–56). The primary reason for ICU Table 1 All cases of active TB (n ¼ 83) Diagnoses
n (%)
Isolated PTB (without overt EPTB)*
46 (55.4)
EPTB* Pleural TB Disseminated TB (multiple sites) or miliary TB TB meningitis TB abdomen TB involving other sites
13 11 7 4 2
* 23 of 37 (62.2%) patients with EPTB had concomitant active PTB. TB ¼ tuberculosis; PTB ¼ pulmonary TB; EPTB ¼ extra-pulmonary TB.
(15.7) (13.3) (8.4) (4.8) (2.4)
826
The International Journal of Tuberculosis and Lung Disease
Table 2
Categorical data: ICU survivors vs. non-survivors (per patient, n ¼ 83)
Parameter Patient demographic and risk Female sex HIV-positive CD4 ,200 cells/mm3, n/N AIDS Diabetes mellitus COPD Anti-tuberculosis treatment on admission TB involvement PTB Isolated PTB EPTB Isolated EPTB Previous PTB Haemoptysis AFB-positive Radiology Cavitation Miliary Diffuse bronchopneumonia Lobar consolidation Pleural effusion Lymph node enlargement alone Normal Complications ARDS Renal failure VAP Septic shock DIC MODS
All (n ¼ 83) n
Survivors (n ¼ 45) n
Non-survivors (n ¼ 38) n
45 44 31/44 27 9 6 32
25 26 15/26 13 5 2 22
20 18 16/18 14 4 4 10
0.89 0.66 5.87 1.46 0.94 2.52 0.37
(0.37–2.11) (0.28–1.57) (1.11–30.95) (0.57–3.61) (0.23–3.79) (0.44–14.64) (0.15–0.95)
1 0.467 0.043* 0.590 1* 0.405* 0.060
69 46 37 14 26 15 32
36 23 22 9 17 9 15
33 23 15 5 9 6 17
1.65 1.47 0.68 0.61 0.51 0.75 1.62
(0.58–6.10) (0.61–3.52) (0.28–1.63) (0.18–1.99) (0.20–1.34) (0.24–2.34) (0.66–3.95)
0.590 0.522 0.262 0.558 0.254 0.841 0.403
12 6 39 10 16 3 5
6 3 17 9 11 2 4
6 3 22 1 5 1 1
1.22 1.20 2.26 0.11 0.47 0.58 0.28
(0.36–4.15) (0.23–6.32) (0.94–5.47) (0.01–0.90) (0.15–1.50) (0.05–6.70) (0.03–2.59)
1 1* 0.108 0.018* 0.307 1* 0.369*
26 31 19 23 15 25
12 15 10 10 8 14
14 16 9 13 7 11
1.66 1.46 1.09 1.82 1.04 0.90
(0.63–4.08) (0.60–3.56) (0.39–3.03) (0.69–4.81) (0.34–3.20) (0.35–2.32)
0.450 0.554 0.920 0.332 0.841 1
OR (95%CI)
P value
* Fisher’s exact test (all other P values calculated using v2 test). ICU ¼ intensive care unit; OR ¼ odds ratio; CI ¼ confidence interval; HIV ¼ human immunodeficiency virus; AIDS ¼ acquired immune-deficiency syndrome; COPD ¼ chronic obstructive pulmonary disease; TB ¼ tuberculosis; PTB ¼ pulmonary TB; EPTB ¼ extra-pulmonary TB; AFB ¼ acid-fast bacilli; ARDS ¼ acute respiratory distress syndrome; VAP ¼ ventilator-acquired pneumonia; DIC ¼ disseminated intravascular coagulation; MODS ¼ multiple organ dysfunction syndrome.
admission was acute respiratory failure in two thirds of patients (n ¼ 56). TB was the sole cause of respiratory failure in 19 patients, most of whom had acute respiratory failure due to massive haemoptysis (n ¼ 14). The clinico-radiological syndrome of community-acquired pneumonia (CAP) was present in 38 (45.8%) patients; half of these presented with septic shock (n ¼ 19). Four of the 56 patients presenting with acute respiratory failure had cardiopulmonary arrest necessitating ICU admission. Other reasons for ICU admission included reduced level of consciousness (n ¼ 7), surgery of the gastrointestinal tract (n ¼ 3) and concomitant disease not directly related to TB (n ¼ 3). ICU mortality was 44.2% (39/86 admissions); only 34/83 patients were alive at the time of hospital discharge (in-hospital mortality 59.0%). The three patients who were readmitted all survived their respective ICU and hospital admissions. Comorbid diseases (COPD, diabetes mellitus or HIV/AIDS) were present in 52 patients (62.7%); only six patients had two comorbidities. Of the 44 patients with HIV co-infection, 18 (40.9%) did not survive: 13/31 patients with a CD4 count ,200 cells/mm3
died (51.6%) vs. 2/13 with a CD4 count 7200 cells/ mm3 (15.4%, P ¼ 0.043). About one third of the patients (34.1%) were on antiretrovirals at the time of admission, and just over half were highly active ART-na¨ıve. Cases with PTB and/or EPTB did not differ significantly with regard to ICU mortality (Table 2). Patients with lobar consolidation were at a lower risk of dying (P ¼ 0.018), whereas a higher APACHE II score tended to have a higher ICU mortality (Table 3). No other radiological or laboratory parameter predicted ICU mortality. There was a significant association between renal failure and in-patient death (Table 4). Moreover, patients who died in hospital had a significantly higher APACHE II score (22.6 6 8.5 vs. 18.1 6 7.4, P ¼ 0.016; Table 5). Thirty-two patients (38.6%) were already started on anti-tuberculosis treatment at the time of admission; the others were diagnosed after admission. The median time until start of treatment for patients who were diagnosed with TB in the ICU was 1.6 days (range 0–17). Tracheal aspirates were obtained from all but three patients; 39.8% were AFB-positive.
TB mortality in the ICU
Table 3
827
Continuous data: ICU survivors vs. non-survivors (per patient, n ¼ 83)
Parameter Age, years APACHE II score CD4, cells/mm3 P:F ratio White cell count, x 109/l Platelet count, x 109/l Haemoglobin, g/dl Serum albumin, g/l C-reactive protein, mg/l Creatinine, lmol/l ALT, U/l s-Glucose, mmol/l
Reference NA NA 600-1500 .300 4.0–11.0 150–400 12.0–15.0* 35–50 ,5 ,90* 5–40 4.4–6.1
All patients (n ¼ 83) mean 6 SD 36.5 20.7 160.2 215.2 13.5 282.8 9.76 25.1 175.2 137.7 133.8 8.80
6 6 6 6 6 6 6 6 6 6 6 6
ICU survivors (n ¼ 45) mean 6 SD
12.9 8.3 163.6 113.2 7.1 165.7 2.41 6.3 94.4 152.0 305.1 4.76
36.8 19.3 198.5 220.0 14.1 307.6 9.52 25.1 168.8 131.6 178.6 8.32
6 6 6 6 6 6 6 6 6 6 6 6
12.6 7.7 180.7 115.5 6.7 175.1 2.55 7.0 93.3 139.4 398.8 4.55
ICU non-survivors (n ¼ 38) mean 6 SD 36.1 22.4 104.8 209.4 12.8 253.4 10.04 25.1 183.0 145.0 84.8 9.35
6 6 6 6 6 6 6 6 6 6 6 6
12.6 8.8 119.1 111.8 7.6 150.9 2.23 5.5 97.0 167.4 138.1 5.00
P value 0.803 0.086 0.060 0.668 0.409 0.138 0.257 0.984 0.531 0.690 0.212 0.330
* For females. ICU ¼ intensive care unit; SD ¼ standard deviation; NA ¼ not available; APACHE II ¼ Acute Physiology and Chronic Health Evaluation II; P:F ratio ¼ arterial partial oxygen pressure to inspired fractional oxygen concentration ratio; ALT ¼alanine aminotransferase.
Mycobacterial cultures obtained from a variety of sites, including tracheal aspirates, were positive in 48 patients (57.8%). Other sources included pleural fluid, cerebrospinal fluid, blood, lymph nodes, ascites, urine and placenta. In culture-negative cases, TB diagnosis was based on a high clinical probability, combined with radiological evidence (n ¼ 14), pleural
fluid analysis (n ¼ 13), cerebrospinal fluid analysis (n ¼ 7) or histology (n ¼ 1). Drug susceptibility testing (DST) for first-line antituberculosis drugs was performed on all positive samples, and DST for RMP was performed on Xpertpositive samples. Two patients had INH monoresistance; one had been previously treated for TB.
Table 4 Categorical data: hospital survivors vs. non-survivors (per patient, n ¼ 83) All (n ¼ 83) n
Survivors (n ¼ 34) n
Non-survivors (n ¼ 49) n
45 44 31/44 27 9 6 32
18 19 11/19 10 2 2 16
27 25 20/25 17 7 4 16
1.09 0.82 2.91 1.28 2.67 1.42 0.55
(0.45–2.62) (0.34–1.98) (0.76–11.09) (0.50–3.28) (0.52–13.71) (0.25–8.24) (0.22–1.34)
1 0.823 0.209 0.791 1* 1* 0.273
69 46 37 14 26 15 32
27 20 15 7 11 7 12
42 26 22 7 15 8 20
1.56 0.79 1.03 0.64 0.92 0.75 1.26
(0.49–4.93) (0.33–1.92) (0.43–2.49) (0.20–2.04) (0.36–2.36) (0.24–2.32) (0.51–3.13)
0.647 0.764 1 0.647 1 0.773 0.653
Radiology Cavitation Miliary Diffuse bronchopneumonia Lobar consolidation Pleural effusion Lymph node enlargement alone Normal
12 6 39 10 16 3 5
5 2 13 6 9 1 3
7 4 26 4 7 2 2
0.97 1.42 1.83 0.41 0.46 0.40 0.44
(0.28–3.35) (0.25–8.24) (0.75–4.45) (0.11–1.60) (0.15–1.40) (0.12–16.13) (0.07–2.79)
1* 1* 0.267 0.304 0.271 1* 0.644*
Complications ARDS Renal failure VAP Septic shock DIC MODS
26 31 19 23 15 25
10 7 6 7 4 8
16 24 13 17 11 17
1.16 3.70 1.69 2.05 2.17 1.73
(0.45–3.01) (1.35–10.09) (0.57–4.99) (0.74–5.67) (0.63–7.51) (0.64–4.63)
0.920 0.016 0.498 0.251 0.340 0.396
Parameter Patient demographic and risk Female sex HIV-positive CD4 ,200 cells/mm3, n/N AIDS Diabetes mellitus COPD Anti-tuberculosis treatment on admission TB involvement PTB Isolated PTB EPTB Isolated EPTB Previous PTB Haemoptysis AFB-positive
OR (95%CI)
P value
* Fisher’s exact test (all other P values calculated using v2 test). ICU ¼ intensive care unit; OR ¼ odds ratio; CI ¼ confidence interval; HIV ¼ human immunodeficiency virus; AIDS ¼ acquired immune-deficiency syndrome; COPD ¼ chronic obstructive pulmonary disease; TB ¼ tuberculosis; PTB ¼ pulmonary TB; EPTB ¼ extra-pulmonary TB; AFB ¼ acid-fast bacilli; ARDS ¼ acute respiratory distress syndrome; VAP ¼ ventilator-acquired pneumonia; DIC ¼ disseminated intravascular coagulation; MODS ¼ multiple organ dysfunction syndrome.
828
The International Journal of Tuberculosis and Lung Disease
Table 5
Continuous data: hospital survivors vs. non-survivors (per patient, n ¼ 83)
Parameter Age, years APACHE II score CD4, cells/mm3 P:F ratio White cell count, x 109/l Platelet count, x 109/l Haemoglobin, g/dl Serum albumin, g/l C-reactive protein, mg/l Creatinine, lmol/l ALT, U/l s-Glucose, mmol/l
Reference NA NA 600-1500 .300 4.0–11.0 150–400 12.0–15.0* 35–50 ,5 ,90* 5–40 4.4–6.1
All patients (n ¼ 83) mean 6 SD 36.5 20.7 160.2 215.2 13.5 282.8 9.76 25.1 175.2 137.7 133.8 8.80
6 6 6 6 6 6 6 6 6 6 6 6
12.9 8.3 163.6 113.2 7.1 165.7 2.41 6.3 94.4 152.0 305.1 4.76
Survivors (n ¼ 34) mean 6 SD 35.2 18.1 175.6 209.3 14.2 321.4 9.94 25.3 162.0 109.6 166.1 8.17
6 6 6 6 6 6 6 6 6 6 6 6
11.6 7.4 135.6 110.0 7.4 177.2 2.39 6.9 86.8 135.2 433.0 4.72
Non-survivors (n ¼ 49) mean 6 SD 37.3 22.6 148.4 219.2 13.1 256.1 9.64 25.0 184.4 157.2 110.4 9.22
6 6 6 6 6 6 6 6 6 6 6 6
13.8 8.5 183.9 116.4 6.9 153.4 2.44 6.0 99.6 162.0 161.9 4.80
P value 0.461 0.016 0.592 0.698 0.480 0.077 0.765 0.866 0.331 0.162 0.456 0.330
* For females. ICU ¼ intensive care unit; SD ¼ standard deviation; NA ¼ not available; APACHE II ¼ Acute Physiology and Chronic Health Evaluation II; P:F ratio ¼ ratio of arterial partial pressure of oxygen to inspired fractional concentration of oxygen; ALT ¼alanine aminotransferase.
Another patient previously treated for TB had resistance to INH and EMB. Most patients (83.1%) were anaemic; half of these had Hb ,9 g/dl. White blood cell (WBC) count was elevated in 56.6% of patients; three had a WBC count ,4 3 109. Hypoalbuminaemia was present in most patients (92.5%). There was a high incidence of renal failure (n ¼ 31, 37.3%), which was associated with an increased risk of dying in hospital. The most common radiological finding was diffuse bronchopneumonia (n ¼ 39, 47.0%), followed by pleural effusion (n ¼ 16, 19.3%). Lobar consolidation (n ¼ 10) was the only radiological parameter statistically significantly associated with ICU survival, but not with in-hospital survival (Table 2). Neither ARDS (n ¼ 26, 35.6%) nor MODS (n ¼ 25, 30.1%) was associated with increased mortality. The ICU course was complicated by the development of VAP in 19 patients (27.7%), which did not adversely influence outcome.
DISCUSSION In this prospective observational study, we found high ICU and in-hospital mortality among patients admitted to the ICU with TB, at around 44% and 59%, respectively. Few clinical parameters, special investigations or other ancillary tests predicted outcome. Only CD4 ,200 cells/mm3 in HIV-co-infected patients (P ¼ 0.043) and the absence of lobar consolidation (P ¼ 0.018) were associated with ICU non-survival, whereas a high APACHE II score (P ¼ 0.016) and the development of renal failure (P ¼ 0.016) were associated with hospital non-survival. Our in-hospital mortality rates of acute respiratory failure secondary to TB requiring mechanical ventilation are in line with other published data, and are much higher than those of patients with respiratory failure due to non-tuberculous pneumonia.8,11–19 They in fact approximate the mortality rate for ARDS,
and are much higher than predicted by the APACHE II score. Based on our mean APACHE II score of 20.7, expected mortality would be 30–40%.22 Although the APACHE II score may underestimate mortality in this patient group,28 we did observe a significant difference in score between survivors and non-survivors. These findings are consistent with previously published retrospective data.8 The ICU mortality observed for patients with PTB and/or EPTB did not differ significantly between groups, nor did smear positivity or the diagnosis of miliary TB influence outcome. Due to the small numbers, we cannot comment on the effect of drug resistance on outcome. Several studies have suggested a link between delays in initiating anti-tuberculosis treatment and mortality.12,14,29 Due to the high TB incidence in our setting, we always had a high index of suspicion, particularly in HIV-positive patients, where presentation is often atypical, with a low threshold for initiating empiric anti-tuberculosis treatment. While 38.6% of patients were already on anti-tuberculosis treatment on admission to the ICU, the median time to start of treatment was only 1.6 days. Acute renal failure is a well described complication of critical illness; it is common in the ICU and associated with high ICU mortality. More than a third of our patients had acute renal failure, some requiring dialysis, which was associated with increased risk of dying; this is in keeping with previously published ICU data.12,15,18 Other complications, such as septic shock, ARDS, MODS and VAP, although common, were not linked to poor outcome, as described previously.11,12,14,18 The white blood cell count in our patient population was elevated in 56.6% of patients, and was much higher than in a Brazilian study conducted in a similar setting.18 This can potentially be explained by the high incidence of CAP syndrome, found in 45.8% of admissions; however, no association with mortality
TB mortality in the ICU
was observed. We also did not find an association between low platelet count, low serum albumin, low Hb level, raised liver enzymes or lymphopaenia and poor outcomes.13,14,29–31 Although early retrospective data showed a detrimental effect of HIV on outcome,32 we found no association between the diagnosis of HIV or AIDS and mortality, in keeping with recent retrospective data from Brazil.18 ICU mortality among the HIVinfected subgroup was 40.9%, which is lower than the overall ICU mortality of 44.2%. A marked difference in mortality was observed when stratifying patients according to CD4 count: 13/31 patients with CD4 count ,200 cells/mm3 died (51.6%) vs. 2/13 with CD4 count 7200 cells/mm3 (15.4%, P ¼ 0.043). Profound decreases in CD4 counts have been reported even in the absence of HIV infection, and caution should be used when interpreting data on CD4 counts in general, and particularly in the acutely ill patient.33 In an ICU study conducted by Feeney et al., of 102 ICU patients, only three of whom were HIV-positive, 41% had CD4 levels ,400 cells/mm3 and 29% had ,300 cells/ mm3.34 Low CD4 cell counts in patients diagnosed with TB and subsequent elevation in CD4 count after initiation of anti-tuberculosis treatment have been described in several case series, suggesting that TB itself has an impact on CD4 cell homeostasis.35,36 Contrary to previously published data, lack of cavitation was not linked to poor outcome.31 On the contrary, consolidation was associated with improved ICU, but not hospital outcome. The less extensive lung involvement in this group, with better arterial partial oxygen pressure to inspired fractional oxygen concentration ratios (P:F ratios), might explain the better outcome, as the most common presentation was diffuse bronchopneumonic infiltrates, indicating diffuse and potentially non-reversible lung damage. TB is known to be a slowly responding disease, but the more localised lung involvement may have resulted in better outcomes. Although retrospective case studies have been published from both low- and high-income countries, there is an absence of prospective data on TB in the ICU. Ours is the first prospective study to be conducted in an area of very high TB prevalence. We were able to enrol a large number of patients over a relatively short time period, ensuring that standard of care remained unchanged during the study period and thus limiting the effect of changes in critical care over the years. The high incidence of TB-HIV coinfection made it possible to compare both HIVinfected and non-HIV-infected patients with TB. The design also enabled us to look at a large number of variables, including incidence of multidrug-resistant TB and both PTB and EPTB, whereas previous retrospective data were either unavailable or incomplete with regard to these variables. All of our
829
patients required mechanical ventilation, whereas in previous studies the number of ventilated patients was variable. Limited ICU resources may have introduced a selection bias toward healthier patients being admitted to the ICU. The tendency to admit healthier patients with a better prognosis to the ICU could account for the lack of significance of certain parameters in predicting mortality. Delays in referral — both before admission to hospital and between the emergency department and the ICU — may also have contributed to a delay in diagnosis and treatment of TB and to the high incidence of acute renal failure and other organ failures. Another potential limitation is that 38.6% of patients were already started on antituberculosis treatment at the time of admission. In conclusion, we found that both ICU and inhospital mortality of TB patients admitted to the ICU was extremely high. Few clinical parameters, special investigations or other ancillary tests predicted outcome. Only CD4 ,200 cells/mm3 in HIV-coinfected patients and absence of lobar consolidation were associated with mortality in the ICU, whereas a high APACHE II score and the development of renal failure were associated with hospital mortality. Conflict of interest: none declared.
References 1 Bock N, Reichman L B. Tuberculosis and HIV/AIDS: epidemiological and clinical aspects (World Perspective). Semin Respir Crit Care Med 2004; 25: 337–345. 2 World Health Organization. Global tuberculosis control. Report 2013. WHO/HTM/TB/2013.11. Geneva, Switzerland: WHO, 2013. http://www.who.int/tb/publications. Accessed March 2014. 3 Western Cape Department of Health HIV/AIDS/STI and TB directorate. Action plan for the enhanced response to improve TB control in the Western Cape. Report 2012. Cape Town, South Africa: DoH, 2012. 4 Corbett E L, Watt C J, Walker N, et al. The growing burden of tuberculosis: global trends and interactions with the HIV pandemic. Arch Int Med 2003; 163: 1009–1021. 5 Joint United Nations Programme on HIV/AIDS. Report on the global AIDS epidemic 2013. UNAIDS/JC2502/1/E. Geneva, Switzerland: UNAIDS, 2013. http://www.unaids.org/en/media/ unaids/contentassets/documents/epidemiology/2013/gr2013/ UNAIDS_Global_Report_2013_en.pdf. Accessed March 2014. 6 Levy H, Kallenbach J M, Feldman C, Thorburn J R, Abramowitz J A. Acute respiratory failure in active tuberculosis. Crit Care Med 1987; 15: 221–225. 7 Bernard G, Artigas A, Carlet, J, et al. The American-European consensus conference on ARDS: definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994; 149: 818. 8 Hayhurst M D, Kussman B D, Potgieter P D. Acute respiratory failure due to pulmonary tuberculosis: a five year analysis in an intensive care unit. S Afr J Crit Care 1994; 10: 4–7. 9 Potgieter P D. The influence of the tuberculosis epidemic on ICU resources. S Afr Med J 1995; 85: 1196. 10 Bateman E D. ICU care for tuberculosis patients—fielding on the boundary. S Afr Med J 1997; 87: 635–637. 11 Lee P L, Jerng J S, Chang Y L, et al. Patient mortality of active
830
12
13
14
15
16
17
18
19
20
21
22
23
24
The International Journal of Tuberculosis and Lung Disease
pulmonary tuberculosis requiring mechanical ventilation. Eur Respir J 2003; 22: 141–147. Penner C, Roberts D, Kunimoto D, Manfreda J, Long R. Tuberculosis as a primary cause of respiratory failure requiring mechanical ventilation. Am J Respir Crit Care Med 1995; 151: 867–872. Frame R N, Johnson M C, Eichenhorn M S, Bower G C, Popovich J. Active tuberculosis in the medical intensive care unit: a 15-year retrospective analysis. Crit Care Med 1987; 15: 1012. Zahar J-R, Azoulay E, Klement E, et al. Delayed treatment contributes to mortality in ICU patients with severe active pulmonary tuberculosis and acute respiratory failure. Intensive Care Med 2001; 27: 513–520. Erbes R, Oettel K, Raffenberg M, Mauch H, Schmidt-Ioanas M, Lode H. Characteristics and outcome of patients with active pulmonary tuberculosis requiring intensive care. Eur Respir J 2006; 27: 1223–1228. Ryu Y J, Koh W J, Kang E H, et al. Prognostic factors in pulmonary tuberculosis requiring mechanical ventilation for acute respiratory failure. Respirology 2007; 12: 406–411. Lin S M, Wang T Y, Liu W T, et al. Predictive factors for mortality among non-HIV infected patients with pulmonary tuberculosis and respiratory failure. Int J Tuberc Lung Dis 2009; 13: 335–340. Silva D R, Menegotto D M, Schulz L F, Gazzana M B, Dalcin P T. Mortality among patients with tuberculosis requiring intensive care: a retrospective cohort study. BMC Infect Dis 2010, 10: 54–61. Alshimemeri A A, Arabi Y M, Al-Jahdali H, Olayan A, Al Harbi O, Memish Z. Clinical presentation and outcome of patients diagnosed with active pulmonary tuberculosis in a large critical care unit. Crit Care Shock 2011; 14: 1–6. Geng E, Kreiswirth B, Burzynski J, Schluger N W. Clinical and radiographic correlates of primary and reactivation tuberculosis: a molecular epidemiology study. JAMA 2005; 293: 2740– 2745. Andreu J, Caceres J, Pallisa E, Martinez-Rodriguez. Radiological manifestations of pulmonary tuberculosis. Eur J Radiol 2004; 51: 139–149. Knaus W A, Draper E A, Wagner D P, Zimmerman J E. APACHE II: a severity of disease classification system. Crit Care Med 1985; 13: 818–829. Koegelenberg C F, Nortje A, Lalla U, et al. The pharmacokinetics of enteral antituberculosis drugs in patients requiring intensive care. S Afr Med J 2013; 103: 394–398. Bone R C, Balk R A, Cerra F B, et al. Definitions for sepsis and
25
26
27
28
29
30
31
32
33
34
35
36
organ failure and guidelines for the use of innovative therapies in sepsis. Chest 1992; 101: 1644–1655. Bellomo R, Ronco C, Kellum J A, Mehta R L, Palevsky P. Acute renal failure – definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 2004; 8: R204– R212. ARDS Definition Task Force; Ranieri V M, Rubenfeld G D, Thompson B T, et al. Acute respiratory distress syndrome: the Berlin Definition. JAMA 2012; 307: 2526–2533. American Thoracic Society. Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Respir Crit Care Med 2005; 171: 388–416. Roodt A, Smith C, Feldman C, Levy H, Kallenbach J M. APACHE II severity of illness score in patients with severe active pulmonary tuberculosis. S Afr J Crit Care 1990; 6: 13– 14. Sacks V, Pendle S. Factors related to in-hospital deaths in patients with tuberculosis. Arch Intern Med 1998; 158: 1916– 1922. Maartens G, Willcox P A, Benatar S R. Miliary Tuberculosis: rapid diagnosis, hematologic abnormalities, and outcome in 109 treated adults. Am J Med 1990; 89: 291–296. Barnes P F, Leedom J M, Chan L S, et al. Predictors of shortterm prognosis in patients with pulmonary tuberculosis. J Infect Dis 1988; 158: 366–371. Stoneburner R, Laroche E, Prevots R, et al. Survival in a cohort of human immunodeficiency virus-infected tuberculosis patients in New York City: implications for the expansion of the AIDS case definition. Arch Intern Med 1992; 152: 2033–2037. Aldrich J, Gross R, Adler M, King K, MacGregor R R, Gluckman S J. The effect of acute severe illness on CD 4þ lymphocyte counts in non-immunocompromised patients. Arch Intern Med 2000; 160: 715–716. Feeney C, Bryzman S, Kong L, Brazil H, Deutsch R, Fritz L C. T-lymphocyte subsets in acute illness. Crit Care Med 1995; 23: 1680–1685. Skogmar S, Schon T, Balcha T T, et al. CD4 cell levels during treatment for tuberculosis (TB) in Ethiopian adults and clinical markers associated with CD4 lymphocytopenia. PLOS ONE 2013; 8: e83270. Andersen A B, Range N S, Changalucha J, et al. CD4 lymphocyte dynamics in Tanzanian pulmonary tuberculosis patients with and without HIV co-infection. BMC Infect Dis 2012; 12: 66.
TB mortality in the ICU
i
RESUME
Il existe peu de donne´ es sur les d´eterminants de la mortalit´e de la tuberculose (TB) dans les services de soins intensifs (ICU). O B J E C T I F : D´ecrire les caract´eristiques influen¸cant le devenir des patients admis en ICU pour une TB n´ecessitant une ventilation assist´ee. S C H E´ M A : Tous les patients admis a` l’ICU d’un hopital ˆ universitaire d’Afrique du Sud avec une TB entre janvier 2012 et mai 2013 ont e´ t´e inclus. La gravit´e de la TB a e´ t´e e´ valu´ee selon le score APACHE II. Nous avons enregistr´e les diagnostics de comorbidit´e, les signes cliniques et radiologiques ainsi que les r´esultats du laboratoire. R E´ S U LT A T S : De 83 patients (age ˆ 36,5 6 12,9 ans, 45 femmes, 44 positifs pour le virus de l’immunod´eficience humaine [VIH]) admis avec une TB pulmonaire (n ¼ 69) et/ou extra-pulmonaire (n ¼ 37), 39 patients sont CONTEXTE :
d´ec´ed´es en ICU (mortalit´e 44,2%) ; 10 autres patients sont d e´ ce´ d e´ s pendant l’hospitalisation (mortalit e´ hospitali`ere globale 59%). Peu de param`etres cliniques ou d’autres examens permettaient de pr´evoir le r´esultat. Seul le comptage des CD4 ,200 cellules/mm3 chez les patients coinfect´es par le VIH (P ¼ 0,043) et l’absence de consolidation lobaire (P ¼ 0,018) e´ taient associ´es a` une mortalit´e en ICU, tandis qu’un score APACHE II e´ lev´e (22,6 contre 18,1 ; P ¼ 0,016) et la survenue d’une insuffisance r´enale aigu¨e (P ¼ 0,016) e´ taient associ´es a` une mortalit´e hospitali`ere. C O N C L U S I O N : La mortalit´e des patients tuberculeux admis a` l’ICU est extrˆemement e´ lev´ee. Tr`es peu de param`etres sont associ´es a` une e´ volution d´efavorable, et aucun param`etre n’a pu pr´esager a` lui seul de la mortalit´e en ICU et en service hospitalier. RESUMEN
Existen pocos datos sobre los factores determinantes de la mortalidad por tuberculosis (TB) en las unidades de cuidado intensivo (ICU). O B J E T I V O: Describir las caracter´ısticas que influyen sobre el desenlace cl´ınico de los pacientes tuberculosos hospitalizados que precisan ventilacion ´ meca´nica. M E´ T O D O: Se incluyeron en el estudio todos los pacientes tuberculosos que ingresaron a la ICU de un hospital universitario en Sura´frica entre enero del 2012 y mayo del 2013. Se evaluo´ la gravedad de la enfermedad mediante la puntuacion ´ APACHE II. Se consignaron los diagnosticos ´ de enfermedades concurrentes, las caracter´ısticas cl´ınicas y radiogra´ficas, las investigaciones de laboratorio y los desenlaces cl´ınicos. R E S U LT A D O S: Ingresaron a la ICU 83 pacientes con TB pulmonar (n ¼ 69) y extrapulmonar o ambas (n ¼ 37) 45 mujeres; 44 pacientes (edad 36,5 6 12,9 anos; ˜ positivos al virus de la inmunodeficiencia humana [VIH]). Treinta y nueve pacientes fallecieron en la M A R C O D E R E F E R E N C I A:
unidad (mortalidad 44,2%); otros 10 pacientes fallecieron durante la hospitalizacion ´ (mortalidad intrahospitalaria 59,0%). Pocos para´ metros, investigaciones especiales u otras pruebas complementarias pronosticaron el desenlace. Solo un recuento de c´elulas CD4 ,200 c´elulas/mm3 en los pacientes coinfectados por el VIH (P ¼ 0,043) y la ausencia de consolidacion ´ lobular (P ¼ 0,018) se asociaron con mortalidad en la unidad; sin embargo, una alta puntuacion ´ APACHE II (22,6 contra 18,1; P ¼ 0,016) y la aparicion ´ de insuficiencia renal (P ¼ 0,016) se asociaron con mortalidad intrahospitalaria. ´ N: La mortalidad de los pacientes con CONCLUSIO diagnostico ´ de TB que ingresan a la ICU es extremadamente alta. Muy pocos criterios se asociaron con este desenlace desfavorable y ninguno de ellos pronostico´ la mortalidad general, en la unidad y durante la hospitalizacion. ´
