Our study demonstrated that both preoperative and postoperative HGS reflected postoperative walking ability and quality of life; however, postoperative HGS had a higher prognostic value with functional outcomes than preoperative HGS. The absolute values of the correlation coefficients of postoperative HGS related with Koval scores at postoperative months 6 and 12 and the correlation coefficients of postoperative HGS related with EQ-5D scores at postoperative months 6 and 12 were higher than those at preoperative HGS analysis. In addition, the rate of low strength was significantly lower at postoperative HGS.
This study reconfirmed the importance of HGS as a prognostic factor of functional outcome in patients with hip fracture. The results are consistent with those from previous studies, confirming the prognostic role of HGS after hip fracture12,13,14,15,17,18,19. However, there was no consensus on the timing of HGS measurement, and measurements were taken at inconsistent times throughout these studies. Several studies evaluated preoperative HGS. Monaco et al. demonstrated that grip strength better predicted short-term functional outcomes in women than appendicular lean mass assessed by dual-energy X-ray absorptiometry12. Álvarez et al. concluded that HGS measured at admission for hip fracture was directly related to functional recovery in older patients19. Selakovic et al. showed that HGS measured preoperatively was associated with Barthel index scores at months 3 and 6 post-fracture15. Furthermore, Wehren et al. found a moderate correlation between grip strength and functional outcomes18. Savino et al. also concluded that high HGS at admission was related to a higher probability of independent walking recovery within 1 year of surgery14. Other studies revealed the relationship between postoperative HGS and functional outcomes. Monaco et al. assessed HGS before rehabilitation and observed the prognostic ability of functional outcome of inpatient rehabilitation and at 6 months follow-up in women with hip fractures13. Beloosesky et al. demonstrated that HGS measured 1 week after surgery was associated with the recovery of walking ability at postoperative month 617.
Only two studies found no relationship between grip strength and functional outcome. Steihaug et al. demonstrated that postoperative HGS was not significantly associated with the preoperative Barthel index16. Their study differed from the current study in the relationship between HGS and ‘pre-injury’ functional status was demonstrated, rather than functional outcome after fracture. Gonzalez-Montalvo et al. also failed to show the association between sarcopenia and short-term functional outcome in patients with hip fracture20. Limited comparisons can be drawn between this study and our present study because they did not assess the direct relationship between HGS and long-term functional outcome, including walking ability and activity of daily living, and their functional outcome was measured only at discharge at an average of 10.1 days, without further follow-up.
With respect to the time of HGS assessement, the current study revealed that postoperative HGS reflected prognosis better than preoperative HGS. The correlation coefficient of postoperative HGS with functional outcome was higher than that of preoperative HGS. In addition, postoperative HGS showed a moderate correlation with postoperative 6-month walking ability and postoperative 6- and 12-month EQ-5D. In contrast, preoperative HGS showed a weak correlation with postoperative 6- and 12-month walking ability and EQ-5D score. Although the low strength group based on preoperative HGS showed a lower walking ability and quality of life only at postoperative month 12, the same group based on postoperative HGS had lower walking abilities at postoperative months 6 and 12 and lower EQ-5D scores at postoperative months 3, 6, and 12. Therefore, we believe that postoperative HGS better reflects functional outcomes after hip fracture and is useful in predicting postoperative 6- and 12-month walking ability and quality of life.
The higher prognostic value of postoperative HGS may be attributed to the following reasons. Muscle mass is maintained during the first 10 days after fracture and begins to decrease thereafter31,32. The time difference between pre- and postoperative HGS was approximately 6 days (< 10 days); thus, the difference in HGS may not have resulted from the change in muscle mass between the measurements. First, preoperative HGS was highly influenced by pain, which would have restricted the maximum force exercised by the patients in the preoperative setting. Moreover, preoperative HGS was measured in the supine position due to hip pain, which is not the standard measurement method advocated by the American Society of Hand Therapists33. Teraoka34 reported that HGS in the supine position was weaker than that in the standing or sitting positions, owing to the influence of gravity. Therefore, we believe that the preoperative HGS could have been underestimated.
Additionally, postoperative 1-year mortality rates in geriatric hip fractures are 16% in South Korea35. One-year mortality in the current study was 2.5%, which is much lower than that of studies in the literature. We suggest that two factors might have attributed to lower mortality rates in our study. This study was a prospective cohort study, excluding patients unable to walk before fracture or delayed fracture; agreeing participants were therefore relatively healthy and active. Second, the current study was performed in a well-organized center for geriatric patients, with all surgeries performed as soon as possible by an experienced orthopedic surgeon. Integrated co-management for elderly patients and experience of the surgeon with hip fracture would reduce hospital mortality.
We also examined the associations between HGS asymmetry and weakness in functional limitations. Our analysis revealed the effect of low strength or weakness on the functional relationship, although none were statistically significant. Our study has therefore been unable to determine whether asymmetry or low strength affects functional limitation before surgery. The reliability of the study may also have been impacted due to the large difference in the number of patients in each group; (1) low strength only (n = 28), (2) asymmetry only (n = 4), (3) asymmetry and low strength (n = 28), and (4) symmetric and normal strength. (n = 19). A follow-up study would include an increase in the number of patients and intentions to study the effect of asymmetry and low strength on functional limitations before and after fracture.
There are several limitations to this study. First, this was a single-center study in a tertiary hospital, with all admitted patients of Asian descent. External validity is therefore required to support the global application of our findings. Second, two patients were lost to follow-up at 12 months, which may have affected the statistical analysis. Additionally, admission to surgery and length of stay intervals varied substantially. Patients with long hospital stays (especially 56 days) were included because of their poor condition and rehabilitation requirements. We confirmed that this did not affect the results, but the reliability may have been impacted. In addition, the functional outcomes in our study differed from those in previous studies. The following modalities were used in previous studies: Barthel index12,13,15,16,19,20; Functional Ambulation Category20; Functional Independence Measure17; Activities of Daily Living18; and Instrumental Activities of Daily Living14. Further studies may be needed to determine the most effective scoring system for measuring functional outcome in patients with hip fracture, or to determine whether the type of scoring system affects the relationship with HGS.
This study also had several strengths. First, we provided postoperative 1-year functional outcomes. Among previous studies, few had follow-up periods of 12 months14,18, of those that did, most were less than 6 months12,13,15,16,17,19. Second, previous studies were limited to one-time measurements of HGS12,13,16,17,19; however, the current study evaluated preoperative and postoperative HGS and determined the most meaningful measurement. To the best of our knowledge, this is the first study to compare the prognostic value of preoperative and postoperative HGS. In addition, we increased the validity of our study by using two indicators, walking ability and quality of life, as the functional outcomes. Our study has clinical importance in that by evaluating the value of HGS as a prognostic factor and providing an appropriate timing of HGS measurement, its value is maximized. In clinical practice, postoperative HGS measurement would be more valuable and convenient for patients and physicians when the patient is stable and in a comfortable environment without pain.