Li-Kheng Ngov, MD

Clinical question: Does delirium contribute to chance of death?

Background: Delirium is a well-recognized predictor of mortality. Prior observational studies have estimated a risk of death two to four times higher in ICU patients with delirium compared with those who do not experience delirium. The degree to which this association reflects a causal relationship is debated.

Study design: Prospective cohort study; used logistic regression and competing risks survival analyses along with a marginal structural model analysis to adjust for both baseline characteristics and severity of illness developing during ICU stay.

Setting: Single ICU in the Netherlands.

Synopsis: Regression analysis of 1,112 ICU patients confirmed the strong association between delirium and mortality; however, additional analysis, adjusting for the severity of illness as it progressed during the ICU stay, attenuated the relationship to nonsignificance. This suggests that both delirium and mortality were being driven by the common underlying illness.

In post hoc analysis, only persistent delirium was associated with a small increase in mortality. Although this observational study can neither prove nor disprove causation, the adjustment for changing severity of illness during the ICU stay was more sophisticated than prior studies. This study suggests that delirium and mortality are likely companions on the road of critical illness but that one may not directly cause the other.

Bottom line: Delirium in the ICU likely does not cause death, but its presence portends increased risk of mortality.

Citations: Klouwenberg PM, Zaal IJ, Spitoni C, et al. The attributable mortality of delirium in critically ill patients: prospective cohort study. BMJ. 2014;349:g6652. Inouye SK, Westendorp RGJ, Saczynski JS. Delirium in elderly people. Lancet. 2014;383(9920):911-922.

Clinical question: Does delirium contribute to chance of death?

Background: Delirium is a well-recognized predictor of mortality. Prior observational studies have estimated a risk of death two to four times higher in ICU patients with delirium compared with those who do not experience delirium. The degree to which this association reflects a causal relationship is debated.

Study design: Prospective cohort study; used logistic regression and competing risks survival analyses along with a marginal structural model analysis to adjust for both baseline characteristics and severity of illness developing during ICU stay.

Setting: Single ICU in the Netherlands.

Synopsis: Regression analysis of 1,112 ICU patients confirmed the strong association between delirium and mortality; however, additional analysis, adjusting for the severity of illness as it progressed during the ICU stay, attenuated the relationship to nonsignificance. This suggests that both delirium and mortality were being driven by the common underlying illness.

In post hoc analysis, only persistent delirium was associated with a small increase in mortality. Although this observational study can neither prove nor disprove causation, the adjustment for changing severity of illness during the ICU stay was more sophisticated than prior studies. This study suggests that delirium and mortality are likely companions on the road of critical illness but that one may not directly cause the other.

Bottom line: Delirium in the ICU likely does not cause death, but its presence portends increased risk of mortality.

Citations: Klouwenberg PM, Zaal IJ, Spitoni C, et al. The attributable mortality of delirium in critically ill patients: prospective cohort study. BMJ. 2014;349:g6652. Inouye SK, Westendorp RGJ, Saczynski JS. Delirium in elderly people. Lancet. 2014;383(9920):911-922.

Clinical question: Does delirium contribute to chance of death?

Background: Delirium is a well-recognized predictor of mortality. Prior observational studies have estimated a risk of death two to four times higher in ICU patients with delirium compared with those who do not experience delirium. The degree to which this association reflects a causal relationship is debated.

Study design: Prospective cohort study; used logistic regression and competing risks survival analyses along with a marginal structural model analysis to adjust for both baseline characteristics and severity of illness developing during ICU stay.

Setting: Single ICU in the Netherlands.

Synopsis: Regression analysis of 1,112 ICU patients confirmed the strong association between delirium and mortality; however, additional analysis, adjusting for the severity of illness as it progressed during the ICU stay, attenuated the relationship to nonsignificance. This suggests that both delirium and mortality were being driven by the common underlying illness.

In post hoc analysis, only persistent delirium was associated with a small increase in mortality. Although this observational study can neither prove nor disprove causation, the adjustment for changing severity of illness during the ICU stay was more sophisticated than prior studies. This study suggests that delirium and mortality are likely companions on the road of critical illness but that one may not directly cause the other.

Bottom line: Delirium in the ICU likely does not cause death, but its presence portends increased risk of mortality.

Citations: Klouwenberg PM, Zaal IJ, Spitoni C, et al. The attributable mortality of delirium in critically ill patients: prospective cohort study. BMJ. 2014;349:g6652. Inouye SK, Westendorp RGJ, Saczynski JS. Delirium in elderly people. Lancet. 2014;383(9920):911-922.

Clinical question: How does perioperative hyperglycemia affect the risk of adverse events in diabetic patients compared to nondiabetic patients?

Background: Perioperative hyperglycemia is associated with increased rates of infection, myocardial infarction, stroke, and death. Recent studies suggest that nondiabetics are more prone to hyperglycemia-related complications than diabetics. This study sought to analyze the effect and mechanism by which nondiabetics may be at increased risk for such complications.

Study design: Retrospective cohort study.

Setting: Fifty-three hospitals in Washington.

Synopsis: Among 40,836 patients who underwent surgery, diabetics had a higher rate of perioperative adverse events overall compared to nondiabetics (12% versus 9%, P<0.001). Perioperative hyperglycemia, defined as blood glucose 180 or greater, was also associated with an increased rate of adverse events. Ironically, this association was more significant in nondiabetic patients (odds ratio, 1.6; 95% CI, 1.3–2.1) than in diabetic patients (odds ratio, 0.8; 95% CI, 0.6–1.0). Although the exact reason for this is unknown, existing theories include the following:

• Diabetics are more apt to receive insulin for perioperative hyperglycemia than nondiabetics (P<0.001);
• Hyperglycemia in diabetics may be a less-reliable marker of surgical stress than in nondiabetics; and
• Diabetics may be better adapted to hyperglycemia than nondiabetics.

Bottom line: Perioperative hyperglycemia leads to an increased risk of adverse events; this relationship is more pronounced in nondiabetic patients than in diabetic patients.

Citation: Kotagal M, Symons RG, Hirsch IB, et al. Perioperative hyperglycemia and risk of adverse events among patients with and without diabetes. Ann Surg. 2015;261(1):97–103.  TH

Visit our website for more physician reviews of hospitalist-focused literature.

Clinical question: How does perioperative hyperglycemia affect the risk of adverse events in diabetic patients compared to nondiabetic patients?

Background: Perioperative hyperglycemia is associated with increased rates of infection, myocardial infarction, stroke, and death. Recent studies suggest that nondiabetics are more prone to hyperglycemia-related complications than diabetics. This study sought to analyze the effect and mechanism by which nondiabetics may be at increased risk for such complications.

Study design: Retrospective cohort study.

Setting: Fifty-three hospitals in Washington.

Synopsis: Among 40,836 patients who underwent surgery, diabetics had a higher rate of perioperative adverse events overall compared to nondiabetics (12% versus 9%, P<0.001). Perioperative hyperglycemia, defined as blood glucose 180 or greater, was also associated with an increased rate of adverse events. Ironically, this association was more significant in nondiabetic patients (odds ratio, 1.6; 95% CI, 1.3–2.1) than in diabetic patients (odds ratio, 0.8; 95% CI, 0.6–1.0). Although the exact reason for this is unknown, existing theories include the following:

• Diabetics are more apt to receive insulin for perioperative hyperglycemia than nondiabetics (P<0.001);
• Hyperglycemia in diabetics may be a less-reliable marker of surgical stress than in nondiabetics; and
• Diabetics may be better adapted to hyperglycemia than nondiabetics.

Bottom line: Perioperative hyperglycemia leads to an increased risk of adverse events; this relationship is more pronounced in nondiabetic patients than in diabetic patients.

Citation: Kotagal M, Symons RG, Hirsch IB, et al. Perioperative hyperglycemia and risk of adverse events among patients with and without diabetes. Ann Surg. 2015;261(1):97–103.  TH

Visit our website for more physician reviews of hospitalist-focused literature.

Clinical question: How does perioperative hyperglycemia affect the risk of adverse events in diabetic patients compared to nondiabetic patients?

Background: Perioperative hyperglycemia is associated with increased rates of infection, myocardial infarction, stroke, and death. Recent studies suggest that nondiabetics are more prone to hyperglycemia-related complications than diabetics. This study sought to analyze the effect and mechanism by which nondiabetics may be at increased risk for such complications.

Study design: Retrospective cohort study.

Setting: Fifty-three hospitals in Washington.

Synopsis: Among 40,836 patients who underwent surgery, diabetics had a higher rate of perioperative adverse events overall compared to nondiabetics (12% versus 9%, P<0.001). Perioperative hyperglycemia, defined as blood glucose 180 or greater, was also associated with an increased rate of adverse events. Ironically, this association was more significant in nondiabetic patients (odds ratio, 1.6; 95% CI, 1.3–2.1) than in diabetic patients (odds ratio, 0.8; 95% CI, 0.6–1.0). Although the exact reason for this is unknown, existing theories include the following:

• Diabetics are more apt to receive insulin for perioperative hyperglycemia than nondiabetics (P<0.001);
• Hyperglycemia in diabetics may be a less-reliable marker of surgical stress than in nondiabetics; and
• Diabetics may be better adapted to hyperglycemia than nondiabetics.

Bottom line: Perioperative hyperglycemia leads to an increased risk of adverse events; this relationship is more pronounced in nondiabetic patients than in diabetic patients.

Citation: Kotagal M, Symons RG, Hirsch IB, et al. Perioperative hyperglycemia and risk of adverse events among patients with and without diabetes. Ann Surg. 2015;261(1):97–103.  TH

Visit our website for more physician reviews of hospitalist-focused literature.