This study documented a prevalence of 77% of swallowing and eating difficulties in an elderly postoperative hip fracture population. Living in a nursing home before hospital admission, a low habitual NMS, and the absence of cardiac comorbidity were found to be significantly associated with swallowing and eating difficulties.
Previous studies have focused on the prevalence of dysphagia following hip fracture surgery and documented a prevalence of 5–54% for patients ≥65 years operated on for hip fracture [39
]. Due to the broader perspective on swallowing and eating difficulties in this study combined with different ways of assessing the difficulties, a direct comparison of results is impossible. However, the V-VST test results can be compared to previous studies. This study documented dysfunction in swallowing regarding effectivity and/or safety assessed by the V-VST in 49% of the patients tested. This is higher than the prevalence of dysphagia documented by Love et al. (2013), who found a prevalence of 34% [41
], but comparable to prevalence of 42% and 54% documented by Meals et al. (2016) and Beric et al. (2019), respectively [39
]. Byun et al. (2019) found a prevalence of dysphagia of 5% [40
]. However, results are not comparable because only patients who were considered at high risk of dysphagia, based on patient history, patient-reported symptoms of dysphagia, and a simple water swallowing test, underwent an assessment in that study.
In this study, living in a nursing home before hospital admission was found to be significantly associated with swallowing and eating difficulties. This supports the findings in previous studies focusing on patients with hip fracture [39
] and on elderly patients in general [8
]. In Denmark, due to the characterization of patients with hip fracture including high age and a high degree of comorbidity, patients living in a nursing home before hospital admission are often discharged from the hospital very shortly after the operation to avoid delirium. There is only a very short time postoperatively in the hospital to focus on swallowing and eating difficulties, and often the patient is discharged before a swallowing and eating assessment is performed. The results in the present study and the previous studies mentioned highlight the importance of caregivers focusing on possible swallowing and eating difficulties in patients living in a nursing home.
Furthermore, a low habitual NMS was found to be significantly associated with swallowing and eating difficulties in this study. NMS is a validated predictor of long-term mortality and rehabilitation outcome in patients with hip fracture [52
]. The finding of an association between swallowing and eating difficulties and a low habitual NMS supports previous findings of an association between eating difficulties and reduced activity of daily living and may indicate the risk of long-term mortality and not optimal outcome of rehabilitation [9
]. Finally, the absence of cardiac comorbidity was found to be significantly associated with swallowing and eating difficulties in this study, which most likely is a random finding caused by the small sample size.
Strengths and Limitations
To our knowledge, no studies have previously focused on swallowing and eating difficulties in a broader perspective in an elderly postoperative hip fracture population. The broader perspective enables a focus on several important prerequisites to swallowing and eating that are highly relevant for the population, for instance their sitting position. Therefore, the broader perspective is the main strength of this study. A further strength is that swallowing and eating assessment was performed in 84% of the patients who fulfilled the inclusion criteria.
The sample size in this study was relatively small, which may have led to type II error. Furthermore, data regarding BMI, ASA score, delirium, habitual NMS, habitual swallowing difficulties, type of anesthesia, and CAS are not complete. Love et al. (2013) demonstrated an association between postoperative dysphagia and postoperative delirium [41
] and Beric et al. (2019) that postoperative confusion predicted dysphagia post-surgery [39
]. Meals et al. (2016) demonstrated that the ASA score was a meaningful predictor of dysphagia [42
]. The fact that we could not observe any association between swallowing and eating difficulties and delirium and the ASA score, respectively, in our study may be explained by the small sample size and the missing data.
As mentioned earlier, patients operated for hip fracture are characterized by high age [26
], and therefore cognitive impairment in this patient group is likely. In Denmark, screening for cognitive impairment during hospitalization is not common practice, and therefore screening for cognitive function was not done in this study, though it would have been relevant.
Due to the cross-sectional design in this study, no detection of causal relationships is possible. Furthermore, because there was no follow-up, possible changes in swallowing and eating difficulties over time, for instance as a result of physical training, were not identified.
Habitual NMS and habitual swallowing difficulties were obtained based on self-reporting from the patient, a relative, or a care assistant, and information bias is, therefore, possible. Simultaneously, an underestimation of the habitual swallowing difficulties is possible since patients were not tested before the operation.
Video fluoroscopy and fiberoptic endoscopic evaluation of swallowing are objective assessments of swallowing function [56
]. It was not possible to use these assessments in our clinical setting. Instead, to examine swallowing difficulties, we used V-VST, which is translated into Danish but not yet validated in Denmark. V-VST was chosen because studies have shown a strong correlation between video fluoroscopy and V-VST [43
] and given the fact that V-VST has been recommended in reviews [57
]. V-VST uses a decrease of oxygen saturation ≥3% to detect silent aspiration, which is not a reliable indicator [59
]. Furthermore, pharyngeal residue is impossible to visualize in a bedside screening but was in this study based on the question of patient experience.
To examine eating difficulties we used MEOF-II. MEOF-II is validated and recommended as a measurement for the performance of a meal [3
]. A study recently published provides support for the reliability and validity of the Danish version of the MEOF II [47
]. MEOF-II is not validated for detecting dysphagia, and the tool has no focus on the viscosity of the food, but the occupational therapist who performed the MEOF-II used food and fluids with different viscosities.
Patients not screened for swallowing and eating difficulties were more likely to live habitually in a nursing home and to be delirious postoperatively. As mentioned earlier, living in a residential aged care facility before hospital admission and presence of postoperative delirium has previously been found to be associated with postoperative dysphagia in elderly patients with hip fracture [39
]. Patients not screened waited, on average, for a longer time from admission to surgery, and more of the patients not screened underwent surgery in general anesthesia. A previous study has shown that waiting time to surgery is correlated with an increased risk of serious adverse events during the hospital stay in patients with hip fracture [60
], adverse events out of which some are previously documented to be associated with dysphagia [41
]. Finally, patients not screened had a lower CAS day 1 after surgery. CAS is a valid tool for evaluating basic mobility in patients with hip fracture [61
], and a lower CAS in patients not screened indicates a less independent and thereby more fragile group than patients screened. All these factors can lead to selection bias, underestimating the prevalence of swallowing and eating difficulties.