In our retrospective study of older hip fracture patients, men had higher crude mortality than women in all time points up to 90 days. In-hospital hypernatremia had predictive value for 90-day mortality.
The gender distribution of hip fractures in our study matches local and international studies, the female incidence being slightly over 70% (Pajulammi et al., 2017; Sing et al., 2023). The overall mortality in our study was similar than those reported in previous hip fracture studies (Haugan et al., 2021). For the 90-day mortality, a large Danish register study reported a result of 16 % at 90 days, which is slightly higher than our crude mortality(Gundel et al., 2020). In our study men had higher mortality than women, which is also well in line with previous hip fracture research. In fact, the difference in the risk of mortality in between the two genders in hip fracture cohorts is well established. Two large population-based cohort studies report increased 30-day (Kristensen et al., 2017) and 1 year mortality (Brit et al., 2018) in men compared to women with similar ratios than in our study. The reasons for this sex-related difference are not well known. The difference in comorbidity does not seem to fully explain this. However, it has been hypothesized that men suffer more greatly from the fracture event, postoperative complications, and common geriatric risk factors such as malnutrition, sarcopenia, and functional impairment. (Brit et al., 2018)
Hypernatremia in hip fracture patients has not been widely studied. In the few studies available, the reported prevalence has been between 1.5 to 3% (Li-Tao Shi et al., 2020; Madsen et al., 2016). Thus, our study adds to the previous research. We found hypernatremia in 8.5 % of subjects on admission and 6.8 % of subjets before discharge. This prevalene is higher than in previous studies. Our results focusing on the in-hospital period highlight the detrimental effect of chronic or uncorrected new-onset hypernatremia before discharge. In general hospital population, hypernatremia on admission is known to increase mortality (Arzhan et al., 2022). Persistent hypernatremia in the acute hospital period has also showed an increasing effect on mortality (Thongprayoon et al., 2021). To our knowledge, only two larger studies have reported hypernatremia in patients undergoing orthopedic surgery, both of which were retrospective studies (Madsen et al., 2016; Mc Causland et al., 2014). Of these two, a Danish registry study of over 7,000 hip fracture patients reported a 1.7 % prevalence of hypernatremia on admission and increased 30-day mortality (15.5% vs 9.6%, p = 0.03) compared to patients with normal serum sdium concentration (Madsen et al., 2016). Mortality was decreased if sodium concentration was normalized, emphasizing diligent perioperative treatment of hypernatremia. In the same study, however, similar decrease in mortality was not seen with persistent hyponatremia.
An Israeli study of 155 hip fracture patients investigated the in-hospital trends of sodium levels in relation to functional and mental status. They reported a trend towards postoperative hypernatremia in patients with decreased prefracture mobility and impaired cognition (Beloosesky et al., 2011). A small Italian study reported a 1.4 % prevalence of hypernatemia on admission but did not investigate mortality (Li-Tao Shi et al., 2020).
It is known that hypernatremia marks a sign of a disruption in the body water balance. Dehydration is the main reason of hypernatremia in the older population. Other causes are sodium loading and excessive renal excretion of electrolyte free water, both of which are rare mechanisms (Oracio et al., 2000). As much as over one third of older patients are known to be in a state of dehydration when admitted to hospital (El-Sharkawy et al., 2015), and every fifth hip fracture patient acquires clinically detectable dehydration in the perioperative period (Zanetti et al., 2022). Geriatric patients have multiple predisposing factors to this electrolyte disorder, and hip fracture surgery brings additional stress to the fluid homeostasis.
Most previous studies have focused on the time of admission to hospital. We are not aware of a previous study focusing on in-hospital electrolyte disturbances of hip fracture patients prior to discharge to a primary health care ward. It is known that acute medical issues at discharge such as desaturation, hypotension and fever predict short- and long-term readmissions (Frandsen et al., 2022; Halm et al., 2003). A large American study found that discharge to inpatient care facilities was associated with post-discharge complications, re-operations and readmissions compared to those discharged to home (Malik et al., 2022). This might reflect the baseline functional status and comorbidities of those needing post-acute care in a hospital setting, rather than the harmful effect of discharge destination care per se.
The strengths of this study include the relatively large and representative study population. Our study focused on a specific group of hip fracture patients, who were transferred to primary health care facilities after the acute postoperative period. Those discharged to their own home or to a care home were excluded from the analyses. Thus, the study population was representative of the largest group among hip fracture patients, and relevant to the study hypothesis. The study population was relatively large, so that analyses yield enough statistical power. The mortality rates and gender differences in our study correspond well to earlier research, thus our data can be treated as representative and reliable for the analysis in question. Our study has also several limitations. Firstly, it was a register-based retrospective study, therefore confounding factors do not allow for direct causative deduction. We could extract only a limited number of data elements from the hospital database. For this reason, we do not know what kind of fluid regimens were prescribed or e.g. what proportion of the patients received blood transfusions during the perioperative period. Another limitation is the lack of knowledge about the exact timing of the laboratory tests taken after admission. As the length of stay varied between patients, there is also marked variation in when the last recorded sodium measurement was taken. Also, if no additional laboratory tests were taken before discharge, the admission test results counted as the last results.
Dysnatremia cannot always be fully corrected before transfer, but proper diagnosis should be made, and treatment protocols prescribed and initiated. As the length of stay in the operating hospital is generally short, the primary care facilities receiving hip fracture patients should also have sufficient resources to treat electrolyte imbalances. Recommended approach to hypernatremia includes calculation of fluid deficit and ongoing losses, and administration of appropriate, usually hypotonic, fluids (Brennan et al., 2021). Specialized care protocols have been developed to manage the multitude of medical issues of hip fracture patients. Orthogeriatric care models represent an advanced and holistic development in the care of fragility fracture patients, strongly recommended for implementation in all hip fracture units (Dreinhöfer et al., 2018). In our study, all patients received standard care at the orthopedic and trauma ward. Nowadays, orthogeriatric care model has been implemented also in the acute care period. We recommend active screening for perioperative hypernatremia as part of an orthogeriatric assessment of hip fracture patients. Local guidelines and discharge checklists help to secure the vulnerable period of discharge.