Increased mortality in women: sex differences in burn outcomes

The findings of this study suggest that women who sustain a burn injury are less likely to survive compared to men. Body size does not appear to modulate this effect, as BMI and BSA did not differ among survivors and non-survivors in either sex. Our results are similar to several other published reports indicating that women are more likely than men to die after burn injury [3, 4, 12, 13, 19, 20].

O’Keefe et al. showed the risk of death for women between the ages of 30 and 59 to be nearly twice that in men of the same age [21]. The differing case fatality rates could not be attributed to any imbalance in risk factors, such as burn size or depth, inhalation injury, or age distribution within the 30- to 59-age bracket, or to any differences in complication incidence, such as acute respiratory distress syndrome (ARDS), sepsis, acute renal failure, or pneumonia. No sex-related differences in mortality were found in the younger or older age groups in the same study. Likewise, Kerby et al. found that women between the ages of 20 and 69 sustaining thermal injury have higher odds of mortality when compared with men; however, no sex differential in mortality odds were found in the younger or older age groups [13]. These differences could not be explained by differences in age, race, TBSA burn, inhalation injury, or the presence of pneumonia. McGwin et al. came to a similar conclusion suggesting that there was no sex differential in mortality risk for burn patients 60 years and older [12].

In modern series, infection and sepsis are the primary reasons for mortality occurring after the initial resuscitation period. Therefore, differential response to infection may be a factor in explaining this sex disparity in survival. Differences in infections and immune response between sexes after a burn injury have been described in the literature. Gregory et al. demonstrated a sex difference in cell-mediated immune responses and interleukin (IL)-6 production after thermal injury. After injury, male mice have a rapid increase in IL-6 production and corresponding suppression of cell-mediated immunity, delayed-type hypersensitivity, and splenocyte proliferative responses within the first 24 h, whereas suppression of cell-mediated immunity in female mice is delayed until day 10, as IL-6 levels increase steadily over time [22]. These differences in immune response may explain the increased risk for women in developing infectious complications that lead to sepsis, multiple-system organ failure, and death [13]. Furthermore, the temporal relationship between post-burn time and immune response may explain the differences in the timing of death between men and women.

There are conflicting reports on the association between sex and survival after sepsis in other fields [9, 23, 24]. Several studies have shown that women tolerate infectious challenge better than men [10, 25] and are less likely to die from sepsis, suggesting a sex difference in the immune response [9]. In response to endotoxemia, van Eijk et al. found that women show a more pronounced innate pro-inflammatory immune response with increased rises in C-reactive protein and tumor necrosis factor-α and less attenuation of norepinephrine sensitivity compared to men [26]. Horton et al. examined sex-related differences in post-burn myocardial inflammation and discovered female burns elicited lower IL-10 levels compared to male burns, correlating with a reduced pro-inflammatory response and a subsequently reduced need for compensatory anti-inflammatory response [27]. Offner et al. show that men have an increased risk of developing major infections following severe injury compared to women; however, mortality rates were found to be similar across both sexes [8]. Pietropaoli et al. and McGwin et al. posit that the female sex is an independent risk factor for mortality in critically ill surgical patients with sepsis [11, 23]. It has also been reported that the female sex is at increased risk for death following hospital-acquired pneumonia [24].

The effects of sex hormones on differences in clinical outcomes have been suggested previously, and explored in a few studies. Increases in IL-6 in female mice following burn injury have been shown to correlate with increased circulating estrogen (E2) levels [28]. Onset of critical illness or trauma has been shown to alter the production of gonadal sex hormones, and circulating concentrations of E2 are enhanced within the first few days following sepsis or thermal injury [29–31]. It is particularly noteworthy that the increased likelihood of death among female burn patients has only been apparent during the reproductive years, according to several studies [12, 13, 21]. It is suggested that the lack of sex differences in mortality among the younger and older age groups reflects the decreased levels of estrogen in younger women and post-menopausal women, negating the increased risk of mortality following thermal injury [13]. However, in the current study, the mean age of non-survivor women was 63 years, with the first quartile at 45 years. In comparison, the median age of male non-survivors was 53 years with the first quartile at 39 years. Therefore, male non-survivors trended younger than their female counterparts and there was no discernible adverse effect on survival of being a premenopausal woman. Jeschke et al. also posit that women exert an attenuated inflammatory and hypermetabolic response to a severe burn compared to men, as female pediatric patients in their study showed higher levels of endogenous anabolic hormones, higher estrogen, attenuated stress hormones, and attenuated inflammatory markers [16]. This decrease is reflected in enhanced muscle protein balance and preservation of lean body mass, which are associated with significantly shortened hospital stays. The results of the current study showed a significantly decreased BUN in women on admission, although this difference disappeared in non-survivors. This further adds to the hypothesis that sex differences in protein catabolism may play a role in burn outcome. Angstwurm et al. suggest that mortality is not dependent on gender but rather correlated with elevated sex hormones across both genders: 17beta-estradiol and testosterone in women and 17beta-estradiol and progesterone in men [32].

Another posited explanation for sex differences in burn outcome relies on the etiology or type of burn being the important confounding variable in the relationship between sex and mortality. For example, Muller et al. described a series in which the majority of female burns were self-inflicted; since self inflicted burns tend to be deeper and more extensive, survival is less likely [3]. Fowler et al. suggest that differences in admission practices, decision-making, and processes of care influencing clinical outcomes for critically ill patients may be a cause for older women having a greater risk of death in the intensive care unit (ICU) and hospital after critical illness [33]. Pietropaoli et al. confirm significant disparities in some aspects of care delivery in their study; women were more likely to have limitations placed on aggressive interventions including resuscitation in case of cardiopulmonary arrest and receive packed red blood cell transfusions while men were more likely to receive invasive mechanical ventilation at admission, deep venous thrombosis prophylaxis, and hemodialysis catheters. These differences, however, could not explain the 10% higher mortality risk in female patients with severe sepsis or septic shock [11]. In the present study, we did not see differences in time to excision and grafting or other parameters of care delivery, suggesting this is less likely to be a source of the differential survival.

Differences in body fat distribution between sexes may be a potential explanation for survival differences. Body fat distribution between sexes and across ethnicities has been studied and found to have consistent patterns. Men are more likely to distribute fat in the central or abdominal fat depots and women in more peripheral areas such as gluteal and femoral subcutaneous fat depots. Lipoprotein A lipase activity was increased in omental and visceral fat but not in subcutaneous fat, implying differential sex metabolic activity based on fat distribution [34–36]. Obesity did not seem to impact overall distribution of fat but showed an increase in fat cell size [35, 36]. While many suggest that body fat distribution and hormone-related receptors are important predictors of fat metabolism and related morbidity and mortality, no clear links or mechanisms have been offered in current literature. Another potential explanation for survival differences between sexes is related to differential activity of brown fat activation and metabolism in relation to hormone secretion between pre-menopausal women and men; however, this is not very well understood and speculative [37].

Limitations of the current study include its limited size and retrospective design. The small sample size decreases statistical power and increases the probability of a type II error where we may have not found statistical significance for outcomes differences between sexes that exist. The small number of women in this study (stemming from the preponderance of male burn patients in technologically advanced countries) precluded any multivariable analysis. Given the disproportionate impact of TBSA, age, and inhalation injury on burn outcome, having only 60 women is too small a study population to perform a meaningful multivariable analysis of the impact of sex on outcome. In addition, the retrospective nature of the study, combined with the lack of information on menopausal symptoms in routine admission documentation for burns, did not allow meaningful analysis of differences in outcomes in pre- vs. post-menopausal women, which is an avenue that deserves further exploration. Further analyses should also be conducted regarding the differences in sepsis and organ complications between men and women following thermal injury. Finally, the lack of outcome difference based on BSA and BMI could mean either that no survival differences based on body fat composition exist or that these variables are not sensitive enough to detect a real difference. As such, we suggest that future studies on body fat composition use bioimpedence analysis or dual-energy X-ray absorptiometry (DEXA) for greater accuracy, instead of estimations such as BMI or BSA.

We present these results as hypothesis-generating rather than a conclusive statement on sex and burn outcomes; studies of larger populations are needed, although recent data from the National Burn Repository support our findings of a subtle trend toward higher mortality in women nationwide (Fig. 2). Based on the trend toward earlier mortality in women, it might be beneficial to evaluate differences in responsiveness to resuscitation between sexes, in addition to investigating cytokines and other hormonal differences that may reveal biochemical reasons underlying the observation of poorer outcome.