The results of this experiment are different from the hypothesis that although the rate of women in the frailty population is much higher than that of men, women are indeed a protective factor against postoperative complications and in-hospital death. This is consistent with the so-called male-female health-survival contradiction survival paradox that women have poor health but less serious comorbidities, leading to longer life expectancy.[13]
In joint replacement, the overall number of frail people gradually increased. The results of this study suggested that the demand to improve the quality of life for frail patients has increased in recent years. The gradual increase in the number and proportion of frail patients needing joint replacements also requires clinical doctors to pay more attention to this population. In this study, the ratio of females to males in frail patients was different from that of the general population requiring joint replacement.[9] The incidence of females with frailty was 1.5-2 times higher than that of men regardless of THA or TKA. Moreover, the rate of CCI ≥ 3 in frail females was higher than that in males. This is consistent with the fact that the health of females is worse than that of males.[14] The higher percentage of female frail patients and the poorer health status of frail females may receive more attention from doctors. This speculative assumption may be supported by the findings that females in the frail population are a protective factor for postoperative pneumonia, although frail women are more likely to have chronic pulmonary disease in TKA and pulmonary circulation disorders in THA. This may also reflect that frail female received more attention from clinicians and consequently had a reduced occurrence of postoperative pneumonia.
For comorbidities, there was indeed a sex difference between frail males and frail females. The prevalence rate of most comorbidities in females was significantly higher than that in males. This makes most frail women seem to have poorer health status than males. However, regardless of hip and knee arthroplasty, severe preoperative comorbidities were less common in frail women than in men. We found that the prevalence of preoperative comorbidities, such as congestive heart failure, liver disease, acute renal failure, or solid tumor without metastasis, in frail women was lower than that in frail men. However, frail women were more likely to have chronic pulmonary disease in TKA and pulmonary circulation disorders in THA. A study showed that among comorbidities, those with the highest independent risk for perioperative mortality were congestive heart failure, metastatic cancer, and renal disease.[15]
For perioperative complications, pulmonary complications such as pneumonia and cardiovascular issues were the most significant factors in increasing the risk for mortality.[16, 17] In our study, for TKA, perioperative complications such as pneumonia and acute heart events in frail males were more common and were significantly different from those in frail females, which accounted for the fact that being female was a protective factor against mortality after TKA. For THA, acute pulmonary edema, pneumonia, and acute heart events were also more common and were significantly different in frail males. Although there was no difference in mortality during hospitalization, it may have an impact on the mortality rate after discharge.
This study also had some limitations, which may affect the generality of our research results. Since this retrospective study is based only on observational data, the documentation may be incomplete in the database. This limitation commonly exists in all research based on large datasets because the quality of the data is closely related to the techniques and methodology used for data point collection and input. Another limitation of large-scale studies is the period of database we choose, we had to choose this period of data because the instrument used to assess frailty in this study is defined by corresponding ICD-9 codes, which were updated to ICD-10-CM in NIS in 2015.[8]