Predictors of survival after severe dysphagic stroke.

BACKGROUND AND PURPOSE: Dysphagia is estimated to occur in up to 50% of the stroke neurorehabilitation population. Those patients with severe neurogenic oropharyngeal dysphagia (NOD) may receive feeding gastrostomy tubes (FGT) if noninvasive therapies prove ineffective in eliminating aspiration or sustaining adequate nutritional intake. Our aim was to quantify the recovery of swallowing function, and to identify variables predictive of survival after dysphagic stroke requiring FGT placement.

METHODS: We identified consecutive stroke patients with severe dysphagic stroke requiring FGT placement admitted to a rehabilitation hospital between May 1998 and October 2001. The medical records were reviewed, and demographic, clinical, videofluoroscopic (VSS) and neuroimaging information were abstracted. A follow-up telephone interview was performed to determine whether the FGT was still in use, had been removed,or if the patient had died. State death certificate records were reviewed to ascertain date of death for subjects who had expired by the time of follow-up. Univariate and multivariate analyses were performed.

RESULTS: 11.6 % (77/664) of stroke patients admitted during the study period had severe dysphagic stroke with FGT insertion. Follow-up was available for 66 (85.7 %) of these individuals at a mean of two years after acute stroke. On follow-up 64% (42/66) of the patients were alive and 45 % had had the FGT removed and resumed oral diets. On univariate analysis patients who were alive at the time of follow-up had received FGT feeding for a shorter period of time (p < 0.0003), showed no signs of aspiration on the Clinical Assessment of Feeding & Swallowing (CAFS,p < 0.020) and on the Videofluoroscopic Swallowing Study (VSS, 0.001), had a better discharge FIM-Score (Functional Independence Measure) for eating (p < 0.0002) and cognitive function (p < 0.002) as well as better discharge FCM-Score (Functional Communication Measure) for swallowing (p < 0.0001). On multivariate analysis we developed a model consisting of FGT removal at discharge from the rehabilitation hospital (p < 0.011) and non-aspiration during VSS (p < 0.040) that was significantly associated with longer survival time during follow-up.

CONCLUSIONS: Severe dysphagia requiring FGT is common in patients with stroke referred for neurorehabilitation. Patients who had a FGT in place at the time of discharge from the stroke rehabilitation unit or aspirated during VSS were substantially more likely to have died by the time of follow-up compared to those who had had the FGT removed and had no signs of aspiration on VSS. However functional outcome measurements (FIM, FCM) including the cognitive function (attention, concentration etc.) could play an important role for prediction of swallowing regeneration and survival in neurorehabilitation. These findings may have practical utility in guiding physicians and speech language pathologists when advising patients and families about prognosis in stroke survivors with severe dysphagia.