記住我
ORIGINAL ARTICLE
Year : 2021 | Volume
: 70
| Issue : 1 | Page : 89-98
Characteristics and outcome determinants in patients with pulmonary tuberculosis in ICU
Gehan M El Assal1, Eman B AbdelFattah1, Mohamed M Nabil2
1 Department of Chest Diseases, Faculty of Medicine, Ain Shams University, Egypt
2 Department of Chest Diseases, Abbassia Chest Hospital, Cairo, Egypt
Correspondence Address:
Eman B AbdelFattah
Associate Professor of Pulmonary Medicine, Chest Department, Ain Shams University; Department of Chest Diseases, Faculty of Medicine, Ain Shams University, Cairo 11843
Egypt
Source of Support: None, Conflict of Interest: None
DOI: 10.4103/ejcdt.ejcdt_35_20
Background Severe tuberculosis requiring intensive care unit care is rare but commonly known to be of bad prognosis. The acute respiratory failure due to pulmonary infections is a common cause for ICU admission. However, despite their high mortality rate, the TB-related critical conditions are rarely reported. Most studies on critical care TB cases were not conducted in high TB burden areas and were retrospective in nature.
Objective To identify potential prognostic factors and determinants of patient outcome in cases with pulmonary TB admitted to ICU.
Subjects and methods This was a prospective observational study conducted on 51 patients with active pulmonary TB who required ICU admission for at least 24 h in ICU unit of Abbassia Chest Hospital during the period from January 1, 2018 till December 31, 2018. All patients were subjected to history and clinical examination, radiological and laboratory workup, and APACHI II score. Patients were diagnosed and managed according to National TB Control Guidelines in Egypt. Hospital course, complications and ICU stay period were monitored and were correlated to the outcome.
Results Mean age of all studied patients was 42.9±16.4 years, 35 male (68.6%) and 16 females. There was Statistically significant correlation between patient outcome, and APACHE II score, arterial blood gases, duration of mechanical ventilation and the total ICU stay in days respectively. ICU complications occurred in 72.5% of patients and they were significantly correlated with patient outcome.
Conclusion Pulmonary TB requiring ICU is rare but markedly of bad prognosis. Patients commonly admitted to ICU due to respiratory failure, sepsis, multi organ failure, ARDS and massive hemoptysis. Bad prognosis was related to respiratory failure, high APACHE II score, need for mechanical ventilation and its duration, ICU stay and related complications. Mechanical ventilation and APACHE II score were found to be predictors of mortality.
Keywords: intensive care unit, pulmonary, tuberculosis
Tuberculosis (TB) is still an important health issue, particularly in developing countries. It is the main infectious cause of death in the world, exceeding HIV/AIDS for the first time [1]. In 1–3% of all TB cases, intensive care is required. Limited number of studies have reported contributing factors to deaths among seriously ill patients with TB, such as extensive fibrocavitary and consolidations on radiographic images, acute respiratory distress syndrome, sepsis, and multiple organ failure, which carries the worst prognosis [2].
ObjectiveThe aim is to identify determinants of patient outcome in cases with pulmonary TB that were admitted to ICU.
Patients and methodsStudy design
This was a prospective observational cohort study of 51 patients with active pulmonary TB who needed to enter the ICU for at least 24 h in Abbassia Chest Hospital, a tertiary care hospital, from January 1, 2018, to December 31, 2018.
Patients were diagnosed and managed according to National TB Control Guidelines in Egypt [3].
Inclusion criteria
Active pulmonary TB diagnosed by either the following:Smear positive for acid-fast bacilli (AFB) for two sputum samples.Positive Gene Xpert MTB/RIF on sputum.Positive TB sputum culture, either solid (Lowenstein–Jensen medium) or liquid (The BD BACTEC MGIT; BD; Becton, Dickinson and Company, Sparks, Maryland, Ireland).Histological pattern of TB granuloma on lung biopsy.Exclusion criteria
The following were the exclusion criteria:
Patients who stay at respiratory ICU for less than 24 h.Extrapulmonary TB.Ethical considerations
This study was submitted for ethical review of the Institutional Review Board (IRB) of Ain Shams University, Faculty of Medicine. All potential patients were invited to join the study through an active consent process and requested to sign a written informed consent. For critically ill patients who cannot sign informed consent, consent was obtained from his/her legally authorized representative.
The following was done for enrolled patients:
Data collection.Demographic data: name, age, sex, and special habits of medical importance (i.e. smoking and drug abuse).Clinical data:Reason for ICU admission.Presenting symptoms and signs.Comorbidities (hypertension, diabetes mellitus, HIV, and other chest diseases).Past medical history (i.e. old TB infection).Relevant medication use (anti-TB drugs).Case definition: new case, relapse, treatment after lost to follow-up (have previously been treated for TB and were declared lost to follow-up at the end of their most recent course of treatment), and treatment failure.Laboratory and related investigations:Arterial blood gases (ABG), complete blood picture, serum sodium and potassium, liver and kidney functions, serum glucose level, coagulation profile (prothrombin time, partial thromboplastin time, and international normalized ratio).Acute Physiology and Chronic Health Evaluation II (APACHE II) score [4].ImagingChest radiograph, posterior–anterior view.Computed tomography chest if needed.Management, course, and complicationsAntituberculous regimen [Cat-I/Cat II/multi-drug resistance (MDR)]/anti-hepatotoxic − according to National TB Control Guidelines in Egypt [3].MDR-TB is known according to patient drug history and drug susceptibility test (Gene Xpert).Oxygen therapy.Mechanical ventilation (MV) if needed.Complications: shock, disseminated intravascular coagulopathy, kidney failure, multiple organ dysfunction syndrome, acute respiratory distress syndrome, and ventilation-acquired pneumonia.ICU stay period.Patients outcomes: respiratory ICU discharge or death.Statistical analysis: descriptive statistics will be made on dichotomous categorical variables, and t tests on continuous data. Unless otherwise indicated, data were given as means and SD.Statistical package: all tests of statistical analysis will be made using the SPSS, version 2010. ResultsThe current study was conducted on 51 cases of pulmonary TB admitted to the ICU for more than 24 h. In the same time interval, 561 cases of pulmonary TB were admitted to the same hospital and did not need ICU. The ratio of patients required ICU to the total cases was 8.33%.
The mean age of the studied cases was 42.9±16.4 years. There were 35 (68.6%) males and 16 (31.4%) females. Regarding race, there were 41 (80.4%) white patients, seven (13.7%) negro patients, and three (5.9%) Asian patients in all studied patients. Description of demographic data and comorbidities is shown in [Table 1]. Descriptive data of past history, diagnosis, and case definition are shown in [Table 2]. There were three cases diagnosed by culture samples (two sputum and one BAL); they were not included in the study because they died before the results were available. Descriptive data of the laboratory data of those patients are shown in [Table 3].
Data describing cause ICU admission, ICU scores, ABG, radiological workup, management, and MV are shown in [Table 4]. Lastly, data describing complications and outcome are shown in [Table 5].
Table 5 Descriptive data about complications, ICU stay, and patient outcomeThere was no statistical significance correlation between demographic data, special habits (age, sex, race, smoking and drug addiction), and comorbidities (diabetes mellitus, cardiac, chronic pulmonary diseases, and HIV) and patient outcome (death or discharge), as shown in [Table 6]. There was no statistically significant correlation between past history, diagnosis, and case definition and patient outcome (death or discharge), as shown in [Table 7]. There was no statistically significant correlation between cause of ICU admission, Glasgow Coma Scale and patient outcome (death or discharge), as shown in [Table 9]. There was no significant correlation between the laboratory data and the patient outcome ([Table 8]).
Table 7 Correlation study between outcome and past history, diagnosis, and case definition
Table 9 Correlation study between outcome and ICU data in studied patientsThe need for MV, duration of connection to ventilator, APACHE II score, and ABG were significantly correlated to patient outcome (P<0.05), as shown in [Table 9]. Presence of complications during the ICU stay and the length of ICU stay both were correlated significantly to patient outcome ([Table 10]).
Table 10 Correlation study between outcome and complications and ICU stayICU stay can be used to discriminate between died patients and discharged patients at a cutoff level of more than 5.5 days (using receiver operating characteristic curve), as shown in [Table 11].
Table 11 Diagnostic performance of ICU stay in discrimination of patient’s outcomeOn studying the different factors that may act as predictors of mortality, MV, and APACHE II score were mortality predictors ([Table 12] and [Figure 1]).
Figure 1 ROC curve between died patients and discharged patients regarding ICU stay. ROC, receiver operating characteristic. Discussion TB is still the most important global mortality cause owing to infectious microorganism. Despite the advances in disease prevention, management, and service delivery, up to 10 million people continue to get infected with TB every year [5].
Around the world, TB remains the single most important cause of dying owing to one infectious organism, and despite improvement of prevention of disease and management, and delivery service, ∼10 million people still contract TB each year.
In the current study, the ratio between TB cases admitted to ICU to the whole number of admitted cases was 8.33%, and this goes in line with the analysis done in Brazil in 2003, which stated that ICU admission was required in 8.5% of cases [6]. Lower ratio was found by Ryu et al. [2] (1–3%). This variation can be explained by many factors, as ICU admission can be influenced not only by respiratory condition (pulmonary TB is seldom the principal cause of acute respiratory failure [7]) but also by comorbidities, level of health care competency and appropriate anti-TB regimen, availability of ICU beds, and guidelines of ICU admission.
Severe TB demanding critical care is rare, but it is of poor prognosis. Most of the previous research studies about critical care tuberculous cases were not conducted in areas with high load of TB and were in retrospective design.
The current study included 51 patients with active pulmonary TB, comprising 35 (68.6%) male patients and 16 (31.4%) females. This was in line with the data from previous studies [8],[9],[10]. Males are more frequent than females. This difference may be owing to greater exposure to droplet infections in male because of outdoor occupation. Other risk factors like smoking, air pollutants exposure, and industrial exposure may play a role.
The mean age of all studied patients was 42.9±16.4 years, with minimum age of 16 years and maximum age 80 years; however, the median age range varies in literature from 41 to 63 years [9],[11].
In the current study, there was no significant correlation between patient outcome and patient sex, the same as Erbes et al. [8]. However, in other study, there was a significant difference between patient outcome (death or discharge) and patient sex, as female sex was a risk factor of mortality but was not a predictor of bad outcome. This discrepancy may be related to difference in disease severity, comorbidities, and need for MV between both studies.
In the current study, 84.3% of patients were new TB cases, 2% were lost to follow-up (previously known as defaulter), and 3.9% were MDR. These findings were in line with study of Mansour et al. [10], which involved only 4% TB resistant cases (two MDR and two extensively drug resistant). In other study [6], previous TB form 23% of cases, 37.5% of which had been discharged, 34.4% were defaulters, 15.6% were receiving treatment when admitted, and 12.5% were admitted for MDR-TB. The difference between the current study and the former two was mainly the prospective nature of the current study and the exclusion of the extrapulmonary TB.
In the current study, we found no statistically significant correlation between patients’ outcome (death or discharge) and smoking or drug addiction. These findings were in line with these studies [10],[12],[13].
In the current study, there was radiological bilateral lung shadow in 30 (58.8%) patients and unilateral lung shadow in 21 (41.2%) patients, with mortality rate of 57.1 and 42.9%, respectively. Thus, radiological extension was not significantly affecting mortality, the same as reported by a previous study [14], where the majority of patient presented with severe radiographic alterations, a possible reflection of a long-lasting disease, though it did not affect mortality significantly.
A history of pulmonary TB was associated with clinical impression of TB, although the radiological picture was not associated with that impression [15].
In the current study, we found no statistically significant correlation between patient outcome and the method of diagnosis. Of 51 patients, 27 (52.9%) were diagnosed by AFB, 16 (31.4%) were diagnosed by Gene Xpert, and eight (15.7%) were diagnosed before admission. One study reported higher death rates associated with patients of pulmonary TB with a positive smear for AFB for mycobacterium TB and with positive cultural examination [16]. However, another one [17] reported the opposite, as they considered smear-positive sputum as a protective factor (less diagnosis delay). It could be said that there was a higher bacterial burden in these patients, maybe denoting a longer pathological process.
In the current study, we found that respiratory failure and the need of MV affect patients’ outcome significantly. A total of 38 patients required MV, and 33 patients died, with mortality rate of ∼94%, which is the same as Penner et al. [18], who found the mortality rate of ∼93% and close to other studies [8],[10],[16],[19]. Similarly, some studies confirmed that the outcome of TB and respiratory failure demanding MV is poor [20],[21].
In the present study, there was a statistically significant correlation between patient outcome and duration of MV, with increase in the mortality rate with increase of days on MV and subsequent mortality. Same finding was seen in the study by Sandrine et al. [16], as MV and need of vasopressor on admission are considered as mortality predictors. Logically, increase duration on MV exposes the patient to serious complications like VAP, sepsis, and pneumothorax.
The most common and significant cause of death in ICU in the present study was ventilatory causes including respiratory failure and MV-related complications, representing 94% of mortality causes, being close to a previous study, which was 83.3% [16].
The current study shows no statistically significant correlation between patient outcome and presence of comorbidities (e.g. diabetes mellitus, cardiac, chronic pulmonary diseases, and HIV). High, but nonsignificant mortality was associated with diabetes (17.1%). The same results were reported in other studies [8],[10],[12],[17],[19].
In fact, death rate in patients with TB-HIV coinfection appears to be associated to the overall patient’s degree of immunosuppression [22]. Patients with HIV infection are more likely to experience extrapulmonary, disseminated, and miliary TB [23], and this type of TB was not included in the current study.
Appropriate antituberculous chemotherapy is an important factor affecting outcome of patients. It was found that a higher mortality is associated with suboptimal treatment, including isoniazid and rifampicin [7],[8]. In other studies also, like in the current study, optimum treatment was difficult to achieve in critically ill patients with TB because of use of alternative regimens (without isoniazid and rifampicin) ascmaintenance therapy in patients with impaired liver function or organ dysfunction and also owing to low serum concentrations of antituberculous drugs as results of uncertain enteral absorption and hypoalbuminemia [7],[24],[25],[26].
In the current study, APACHE II score was significantly correlated to higher mortality; the same was reported by other studies [10],[14],[16].
APACHE II score may underestimate the mortality rate of tuberculous cases requiring MV. The death rate was found to be higher than expected for a given APACHE II score value in such patients in many studies [2],[17],[21].
Regarding the cause of ICU admission, the present study shows that 27 (52.9%) patients were admitted owing to respiratory causes, most commonly owing to respiratory failure; nine (17.6%) patients were admitted owing to metabolic causes and 15 (29.4%) patients were admitted due to both respiratory and metabolic causes. These findings are in the same line with Lin et al. [21], who found that the main cause of ICU admission was respiratory failure, and APACHE II score ranged from 13 to 23 in most studies.
The current study as any other study has its own limitations that should be noted. Some patients died before test results were available, so they were not included in the study. Patients with very bad general condition were difficult to diagnose with rapid deterioration. So these patients were not represented well enough in the study. A single-center study with relatively small size of sample may lack sufficient statistical strength to detect some correlation factors. Despite these limitations, the present results give important implications for similar demographic regions and clinical settings.
ConclusionPulmonary TB requiring ICU is rare but markedly of bad prognosis. Patients are commonly admitted to ICU owing to respiratory causes. Bad prognosis was related to respiratory failure, high APACHE II score, MV and its duration, ICU stay, and related complications. MV and APACHE II score were found to be predictors of mortality.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References
留言 (0)