Clinical profile and outcome of patients intubated in view of altered sensorium- a prospective observational study from North India (2024)

250 mechanically ventilated patients were screened, with 100 fulfilling the inclusion/exclusion criteria. All patients could be followed up till death or discharge. The mean age of the patients was 43 years (SD 16.9) and 55% patients were female. 58% of patients had comorbid illnesses, with diabetes mellitus being the most common (19%). The median CCI Score was 1 (0,2) among survivors with and 1 (0,3) in non-survivors.

The most common cause of altered mental status leading to intubation was meningoencephalitis (34%), followed by septic encephalopathy (17%) and a mixture of various etiologies (16%). Other causes included cerebrovascular accidents, uremic encephalopathy, hepatic encephalopathy, electrolyte imbalances, and seizures. (Table1)

The most frequent Glasgow Coma Scale (GCS) at intubation was 8 (37%), followed by 3 (23%). The median GCS at intubation was higher among survivors (7, 4–8) than non-survivors (6, 4–8), though, not statistically different. (Supplementary Fig.1) Most patients were intubated in the emergency ward (65%), with a smaller percentage in the medicine ward (30%) and ICU (5%). At the time of intubation, patients had a mean systolic blood pressure of 129 mmHg and diastolic blood pressure of 77 mmHg, with survivors (mean 137/81 mmHg) having higher than non-survivors (mean 124/74 mmHg), the difference didn’t reach statistical significance. Patients who survived had higher hemoglobin levels (10.6 vs. 9.4 g/dl, p 0.02) and lower total leukocyte counts (10,800 vs. 13,720/mm³, p 0.03) compared to non-survivors. Patients who died had significantly higher C-reactive protein (CRP) (125 vs 61.5 mg/L, p = 0.02), and lactate dehydrogenase (LDH) (429 vs 349 U/L, p = 0.02), at the time of intubation. Survivors had lower APACHE (21 (16,28) vs. 24 (19,24), p 0.58) and SOFA (7 (5, 9) vs. 8 (6, 10), p 0.062) scores compared to non-survivors, although the difference was not statistically significant. (Table2)

During the hospital stay, 47% of patients developed hospital-acquired infections, and 50% required new-onset vasopressor need. Acute respiratory distress syndrome (ARDS) and liver failure were observed in a small number of patients, and all died. Additionally, 34% of patients required renal replacement therapy, deep vein thrombosis was seen in 5% of patients, but pulmonary embolism (1%) was rare. 32 patients could be successfully extubated, however, 17 (53%) patients needed re-intubation. 4 patients self-extubated, and 3 were successfully discharged. 13 patients were tracheostomized due to persistent altered mental status.

The median length of hospital stay was 11 days, with IQR of 5 to 21.5 days. Patients who survived had a significantly longer duration of hospital stay, a median of 16 days (7, 23) vs. 9 days (4, 20); p = 0.04. Among patients who survived, the median duration of mechanical ventilation was 3 days (2,6 days), significantly less than non-survivors with 7 days (3, 15); p = 0.019. (Supplementary table 1) Overall mortality among all patients intubated due to altered mental status was 65% (+- 4.76%). The highest mortality was among patients intubated due to hepatic encephalopathy (100%, n = 5), followed by intracranial hemorrhage (88%, n = 8). Among the 35 patients discharged, the median mRS scale was 4 (2, 5), with 9 (25.7%) having bedsores at discharge. However, the maximum number of patients discharged i.e., 12 (34.3%) had an mRS of 5.

Cut-off points were determined for APACHE II (22), SOFA (7), CRP (83 mg/L), and LDH (470U/l) levels using ROC curve analysis. (Supplementary Fig.2, Supplementary table 2) Univariable regression analysis showed higher age, abnormal TLC, increased CRP, LDH, APACHE II score, and in-patient intubation were associated with increased mortality. The cause of intubation was also predictive of mortality i.e., patients intubated due to reasons such as seizures, and electrolyte imbalance constituting the ‘Others’ group had protective odds. GCS at intubation was not predictive of mortality.

Multivariable analysis was done with a model consisting of age > 30 years, CCI score of 1 or more, cause of intubation, in-patient intubation, GCS at intubation < 6, albumin < 3.5 (g/dl), CRP > 83 (mg/L), LDH > 470 (IU/l) APACHE II score > 22 and SOFA score > 7. The adjusted odds ratio was calculated, high CRP, high LDH, and in-patient intubation were independently associated with mortality. (Table3)

SOFA and APACHE II scores underestimated mortality with predicted mortality rates of 40% and 50%. Higher mortality was seen at lower SOFA and APACHE II scores, than are predicted. (Supplementary Figs.3,4) Patients across all SOFA and APACH II scores had high inflammatory markers. (Supplementary Figs.5,6)