Does metformin treatment reduce cancer-associated mortality in type 2 diabetes mellitus (DM), is there an association between type 2 diabetes mellitus and postoperative pneumonia, and how does diabetes affect morbidity and mortality in COVID-19?

Updated: Sep 27, 2021
  • Author: Romesh Khardori, MD, PhD, FACP; Chief Editor: George T Griffing, MD  more...
  • Print

In a meta-analysis of 20 publications comprising 13,008 cancer patients with concurrent type 2 diabetes, researchers found that patients treated with metformin had better overall and cancer-specific survival than those treated with other types of glucose-lowering agents. [107, 108] These improvements were observed across cancer subtypes and geographic locations. Risk reduction was significant among patients with prostate, pancreatic, breast, colorectal and other cancers, but not for those with lung cancer. However, it remains unclear whether metformin can modulate clinical outcomes in cancer patients with diabetes.

A study by López-de-Andrés et al found the incidence of postoperative pneumonia in patients with type 2 diabetes to be 21% higher than in nondiabetic patients, although the risk of inhospital mortality following the development of postoperative pneumonia was no greater in the presence of type 2 diabetes. [109]


A retrospective study by Chen et al of 136 COVID-19 patients with diabetes (primarily type 2 diabetes) found that older age, elevated C-reactive protein, and insulin use were risk factors for mortality. The adjusted odds ratio (OR) for mortality in insulin use was 3.58. It has been questioned, however, whether insulin itself is a risk factor or if the increased mortality reflected the characteristics of the patients taking it. [110, 111]

A study by Bode et al indicated that among patients with COVID-19, the US in-hospital death rate for individuals living with diabetes, patients with an HbA1c of 6.5% or higher, and those with hyperglycemia throughout their stay is 29%, a figure over four times greater than that for patients without diabetes or hyperglycemia. Moreover, the in-hospital death rate for patients with no evidence of preadmission diabetes who develop hyperglycemia while admitted was found to be seven times higher (42%). [112, 113]

A whole-population study from the United Kingdom reported that the risk of in-hospital death for patients with COVID-19 was 2.0 times greater for those with type 2 diabetes and 3.5 times higher for individuals with type 1 diabetes. However, patients under age 40 years with either type of diabetes were at extremely low risk for death. [114, 115]

A retrospective study by Zhu et al found that among individuals with COVID-19, those who also had type 2 diabetes mellitus had a mortality rate of 7.8% (versus 2.7% for those without diabetes), as well as a higher rate of multiple organ injury. However, the investigators also reported that among the patients with type 2 diabetes, the mortality rate was lower in those who, during hospitalization, had well-controlled blood glucose, that is, patients with a glycemic variability within 3.9 to 10.0 mmol/L, than in those with poorly controlled blood glucose, in which the upper limit of glycemic variability extended beyond 10.0 mmol/L. [116, 117]

The aforementioned study by Barrera et al indicated that among COVID-19 patients with diabetes, the unadjusted relative risk for admission to an intensive care unit (ICU) is 1.96, and for mortality, 2.78. [38, 39]

Another study from the United Kingdom found that risk factors for mortality in COVID-19 patients with type 1 or type 2 diabetes include male sex, older age, renal impairment, non-White ethnicity, socioeconomic deprivation, and previous stroke and heart failure. Moreover, patients with type 1 or type 2 diabetes had a significantly greater mortality risk with an HbA1c level of 86 mmol/mol or above, compared with persons with an HbA1c level of 48-53 mmol/mol. In addition, an HbA1c of 59 mmol/mol or higher in patients with type 2 diabetes increased the risk as well. The study also found that in both types of diabetes, BMI had a U-shaped relationship with death, the mortality risk being increased in lower BMI and higher BMI but being reduced between these (25.0-29.9 kg/m2). [118, 115]

A literature review by Schlesinger et al strengthened the association between severe diabetes and COVID-19–related mortality, finding that among study patients with diabetes, the likelihood of death from COVID-19 was 75% greater in chronic insulin users. The study also indicated that the chance of death from COVID-19 is 50% less in individuals undergoing metformin therapy than in other patients with diabetes. The investigators suggested that the medications themselves did not impact survival but were indicators of the severity of diabetes in each group, with the prognosis being poorer among those with more severe diabetes. [119, 120]

A retrospective study by Wang et al indicated that hyperglycemia, even in the absence of diabetes, is an independent predictor of 28-day mortality in patients with COVID-19. The investigators reported that on admission to two hospitals in Wuhan, China, 29.1% of study patients with COVID-19 and no prior diagnosis of diabetes had a fasting blood glucose of at least 7.0 mmol/L. It was believed that the individuals with hyperglycemia included not only persons with undiagnosed diabetes, but also nondiabetic patients with acute stress hyperglycemia. With regard to 28-day mortality, it was determined that the hazard ratio in patients with a fasting blood glucose of 7.0 mmol/L or higher was 2.30. [121, 122]

Similarly, another report found that in study patients with COVID-19 who had a blood glucose level of over 6.1 mmol/L, the risk of disease progression was 58% greater, with the mortality risk being 3.22-fold higher. [123]

A retrospective, multicenter study by Carrasco-Sánchez et al supported these results, indicating that among noncritical patients with COVID-19, the presence of hyperglycemia on hospital admission independently predicts progression to critical status, as well as death, whether or not the patient has diabetes. The in-hospital mortality rate in persons with a blood glucose level of higher than 180 mg/dL was 41.1%, compared with 15.7% for those with a level below 140 mg/dL. Moreover, the need for ventilation and intensive care unit admission were also greater in the presence of hyperglycemia. The report involved over 11,000 patients with confirmed COVID-19, only about 19% of whom had diabetes. [124, 125]

In contrast to the above research, a report by Klonoff et al on over 1500 US patients with COVID-19 found no association between hyperglycemia on hospital admission and mortality, in non-ICU patients. However, the in-hospital mortality rate was significantly greater in such patients if they had a blood glucose level above 13.88 mmol/L on the second or third hospital day, compared with those with a level below 7.77 mmol/L. Findings for patients admitted directly to the ICU differed from these, with the investigators determining that mortality was associated with the presence of hyperglycemia on admission but was not significantly linked with a high glucose level on the second hospital day. [126, 127]

A study by Sardu et al indicated that in hospitalized patients with COVID-19 and moderately severe pneumonia, those with diabetes and those who are hyperglycemic are at higher risk of severe disease than are normoglycemic patients without diabetes. Moreover, among the patients in the study with hyperglycemia, the risk of severe disease was lower in those who were treated with insulin infusion, providing further evidence of the importance of in-hospital glucose control. [111, 128]

A study by Cariou et al reported that in patients with diabetes hospitalized for COVID-19, a positive, independent association was found between higher body mass index (BMI) and risk of tracheal intubation and/or death within 7 days. The median BMI in patients who suffered this outcome was 29.1 kg/m2, compared with 28.1 kg/m2 in those who did not. However, an association was not found between long-term glucose control and 7-day tracheal intubation and/or death. Regarding specific outcome rates, the study, in which 88.5% of the diabetes cases were type 2 diabetes, reported that 20.3% of the patients with diabetes who were hospitalized with COVID-19 underwent tracheal intubation within 7 days, while 10.6% died within this time. [129, 130]

A French study, by Wargny et al, indicated that among patients with diabetes who are hospitalized with COVID-19, approximately 20% will die within 28 days. Individuals particularly at risk for mortality over this 4-week period include patients of advanced age, as well as those with a history of microvascular complications (especially those who have had kidney or eye damage), who have dyspnea on admission or inflammatory markers (increased white blood cell [WBC] count, raised C-reactive protein, elevated aspartate transaminase), or who have undergone routine insulin and statin treatment. It should be kept in mind, however, that the data was gathered between March 10 and April 10, 2020, with a statement from Diabetes UK explaining that in people with diabetes, COVID-19–associated mortality has decreased over time as treatment has improved. [131, 132]

The Centers for Disease Control and Prevention (CDC) includes type 2 diabetes in the list of conditions that increase the likelihood of severe illness in persons with COVID-19, and type 1 diabetes in the list of conditions that may increase this likelihood. [133]

Did this answer your question?
Additional feedback? (Optional)
Thank you for your feedback!