A High Risk of Sleep Apnea Is Associated With Less Postoperative Cognitive Dysfunction After Intravenous Anesthesia

Results of an Observational Pilot Study

Soeren Wagner; Joerg Quente; Sven Staedtler; Katharina Koch; Tanja Richter-Schmidinger; Johannes Kornhuber; Harald Ihmsen; Juergen Schuettler


BMC Anesthesiol. 2018;18(139) 

In This Article


It was the aim of this study to evaluate the impact of a high risk of OSAS on the postoperative cognitive dysfunction after intravenous anesthesia. Patients susceptible for a high risk of OSAS were identified using the STOP BANG questionnaire, which is recommended for screening patients with moderate to severe risk for OSAS.[20–22] Recently, Kim and co-workers published a very high sensitivity of 97% for the STOP BANG questionnaire.[23] However, its specificity was low (19%). More important for our purpose was the finding that the positive predictive value was quite high with 86%. Thus, this test seems to be able to detect patients with an at least moderate risk for OSAS. Moreover, several other recently published validation studies revealed fairly good positive predictive values to identify patients with sleep breathing disorder.[24–26] Furthermore, in a systematic review prepared as a part of the Society of Anaesthesia and Sleep Medicine (SASM) publication for the Preoperative Assessment of Patients with Sleep Disordered-Breathing, Opperer and co-workers analyzed how the diagnosis OSA was elucidated. Based on the SASM consensus group criteria for OSAS, 17 publications referred to the diagnosis via polysomnographic results and 15 published articles on screening questionnaires.[27] Chung and co-workers reported that a STOP-Bang score of 5–8 identified patients with a high probability of moderate or severe OSAS in a surgical population.[22] Therefore, we assumed that the patients in our OSAS group were highly suspected of having intermittent hypoxemic episodes.

To account for possible interindividual variation with respect to education levels, we focused on the change in test results prior to and after surgery between OSAS and control patients. In most tests performed, we found significant differences in the change of memory function between OSAS patients and control group patients. OSAS patients in general showed a significantly less decrease or even an increase of the performance compared to the control group.

In this study six separate cognitive test were used to detect different aspects of cognitive impairment as described above. While some tests address a single cognitive function, like the Digit Span test in terms of memory or the RMBT in terms of several aspects of memory, other tests address multiple aspects of cognitive function. Gagnon and colleagues summarized that OSAS patients elicited in all cognitive functions an impaired ability compared to controls.[28] Focusing on the attention aspect, the DemTect and SKT in our study elicited a decrease in attention performance. Moreover, executive function was tested with both tests, DemTect and SKT. Executive function is a complex system of skills including behavioral inhibition, mental flexibility, and working memory. We found impairments in executive function in the preoperative testing phase in OSAS patients compared to controls as described by others.[28] Focusing on executive function, which was tested separately by the FWIT, the OSAS group presented diminished results in our study preoperatively. Working memory was tested separately by the Digit Span test, which likewise showed impaired results in the OSAS group preoperatively. Psychomotor speed was analyzed via TMT und ZVT which revealed decreased results in OSAS patients. Memory function was evaluated using the RMBT, in which the OSAS subjects elicited decreased results in comparison to controls. These results of the present study are in accordance with previously reported findings.[28]

Postoperatively, the DemTect revealed a decrease in verbal memory performance in control patients but not in OSAS patients. Using the RBMT, verbal function was also found to be impaired significantly in control patients, whereas it even improved significantly in the OSAS group. Working memory and memory capacity as assessed by the Digit Span Test were also characterized by an impairment in controls but an improvement in OSAS patients. A similar aspect of memory function is the quality of attention capability, as rated by the SKT test. Again, the test performance of the OSAS group improved after surgery, whereas the performance of the control patients clearly deteriorated. On the other hand, the Color Word Interference Test, which investigates the ability of nomination, alertness and selectivity or interference respectively executive function, did not reveal significant differences between both groups. Moreover, the ZVT did not detect any differences between both groups representing liquid intelligence or processing speed. Finally, using the TMT, which aimed at detecting processing speed as well as rating executive function, we found only minor differences between both groups' test performance.

It is well believed that OSAS patients have impaired cognitive function.[29] There is, however, controversy regarding the grade of impairment. As there is a wide variety in neurocognitive test systems it is difficult to compare the results.[30] The patients in our OSAS group showed indeed a worse preoperative performance in the Demtect test. However, these patients stayed at this level whereas the patients in the control group showed a significantly reduced test performance postoperatively than reported in the literature.[31] Regarding memory function ability, we used the Rivermead Behavioural Memory Test to focus on logical and episodic memory function. In the literature, memory function has been assessed differently and different aspects of memory function are emphasized. While some reviews found impairments in short-term memory function,[32] others reported deficits in verbal and visual delayed long-term memory in OSAS patients.[33] On the other side, no effect was found analyzing the severity of OSAS and its influence on memory function.[34] We found that individuals with a high risk of OSAS performed significantly better in a combination of logical and episodic test settings than control subjects. We concentrated on verbal memory using the subtest "story" in its immediate and delayed version. While we found an increase in verbal memory performance between both test intervals, others described a decrease regarding verbal episodic memory tasks.[35]

The observation that the OSAS patients in our study generally showed a less postoperative decrease or even an increase of the cognitive function might be explained by the phenomenon of hypoxic preconditioning. It is well known that OSAS patients show nocturnal oxygen desaturation.[2] Affected patients present significant consequences including excessive daytime sleepiness. Experimental studies in healthy adults with sleep deprivation lead to the assumption that sleepiness negatively influences neurocognitive performance.[36,37] Others reported that sleep deprivation tolerance in young adults might compensate alteration of cognitive capabilities.[38] Further investigations demonstrated the existence of an ample cognitive reserve and cerebral recruitment enabling individuals to maintain cognitive performance and to resist sleep deprivation induced impairment.[39,40] Functional imaging studies confirmed this protective mechanism in OSAS patients.[41] This observation might explain the finding that patients affected by mild or moderate OSAS can present a normal cognitive function.[42] On the other hand, the severity of hypoxemia altered processing and motor speed performance.[42] It is well known that severe unrecognized and unanticipated desaturation phases may occur following general anesthesia.[5] In both study groups we detected oxygen desaturation after the surgical intervention (Table 1). While our postoperative observations mostly revealed a single hypoxic stimulus following anesthesia randomly, OSAS patients are exposed to multiple hypoxic stimuli per night. Given the assumption that OSAS patients might have adapted to those numerous desaturation events, a single, fortuitous ischemic strike might not negatively affect the neurocognitive performance in these patients. Beneficial effects of prior episodes of ischemia on brain function were found in patients with transient ischemic attack.[43] Hoth and co-workers reported that subjects with a more pronounced hypoxemia performed better in learning and memory tasks compared to those with less severe levels of hypoxemia.[44] These findings led to the assumption that ischemic preconditioning might have protective effects on brain function. Consequently, remote preconditioning approaches have been proposed.[45,46] However, a recently published study failed to prove beneficial effects of remote ischemic preconditioning on postoperative neurocognitive dysfunction in patients who underwent cardiac surgery.[47] On the other hand, lately published data suggest that remote ischemic preconditioning prevent a short-term postoperative cognitive function decline after cardiac surgery.[48] Moreover, He and co-workers published recently that remote ischemic preconditioning improved cognitive function in elderly patients.[49] Already in 2013, Schega and colleagues assumed that additional intermittent hypoxic training and physical exercise enhance cognitive function and quality of life in elderly.[50] Recently in a review published by Li and co-workers, it was summarized that ischemic preconditioning techniques could prevent organ damage and play a neuroprotective role.[51] Therefore, we assume that endogenous hypoxic training might be a favorable cerebral preconditioning factor in OSAS patients and leads to clinical relevance in future.

However, there are some study limitations that should be paid attention to. We merely screened the risk for an OSAS which leads to apnea phases while sleep. Patients have not been diagnosed for OSAS using a sleep laboratory tests or polysomnographic studies. However, the STOP BANG test is an international well established screening tool to quantify the risk for having the OSAS. Although patients were comparable regarding surgical procedures, surgery duration time and severity, they were different in weight and body mass index due to the risk factors for OSAS. The patients of the two groups were of similar age, but the gender distributions were different with a significant higher proportion of males in the OSAS group. However, a two-way ANOVA with gender and group as factors revealed that the observed differences in the change of the cognitive function could be attributed to the factor group whereas the gender did not show a significant effect. Moreover, the study focused on non-cardiac surgery patients only. As O'Brien and colleges summarized that intraoperative hypotension is an intrinsic risk factor of postoperative cognitive dysfunction,[52] intraoperatively hemodynamic stability was considered conscientiously in this study. We did not observe hypotension, defined as systolic blood pressure values below 80 mmHg.[3] However, brief hypotension phases between the measurement intervals might remain undetected and even small periods of hypotension are clearly associated with worse cognitive function in the postoperative phase. Furthermore it is notable that the heterogeneous results reported in literature could be explained by the various severity of OSAS and the wide range in subjects' intellectual function and education. Lastly, this study included only total intravenous anesthesia to avoid an additional confounding factor. However, volatile anesthesia should also be tested and analyzed as cognitive impairment is possible in both general anesthesia regimens.