Effect of age on intraoperative cerebrovascular autoregulation and near-infrared spectroscopy-derived cerebral oxygenation

BACKGROUND

Age is an important risk factor for perioperative cerebral complications such as stroke, postoperative cognitive dysfunction, and delirium. We explored the hypothesis that intraoperative cerebrovascular autoregulation is less efficient and brain tissue oxygenation lower in elderly patients, thus, increasing the vulnerability of elderly brains to systemic insults such as hypotension.

METHODS

We monitored intraoperative cerebral perfusion in 50 patients aged 18-40 and 77 patients >65 yr at two Swiss university hospitals. Mean arterial pressure (MAP) was measured continuously using a plethysmographic method. An index of cerebrovascular autoregulation (Mx) was calculated based on changes in transcranial Doppler flow velocity due to changes in MAP. Cerebral oxygenation was assessed by the tissue oxygenation index (TOI) using near-infrared spectroscopy. End-tidal CO₂, O₂, and sevoflurane concentrations and peripheral oxygen saturation were recorded continuously. Standardized anaesthesia was administered in all patients (thiopental, sevoflurane, fentanyl, atracurium).

RESULTS

Autoregulation was less efficient in patients aged >65 yr [by 0.10 (se 0.04; P=0.020)] in a multivariable linear regression analysis. This difference was not attributable to differences in MAP, end-tidal CO₂, or higher doses of sevoflurane. TOI was not significantly associated with age, sevoflurane dose, or Mx but increased with increasing flow velocity [by 0.09 (se 0.04; P=0.028)] and increasing MAP [by 0.11 (se 0.05; P=0.043)].

CONCLUSIONS

Our results do not support the hypothesis that older patients' brains are more vulnerable to systemic insults. The difference of autoregulation between the two groups was small and most likely clinically insignificant.

Clinical assessment of cerebrospinal fluid dynamics in hydrocephalus. Guide to interpretation based on observational study

OBJECTIVES

The term hydrocephalus encompasses a range of disorders characterised by clinical symptoms, abnormal brain imaging and derangement of cerebrospinal fluid (CSF) dynamics. The ability to elucidate which patients would benefit from CSF diversion (a shunt or third ventriculostomy) is often unclear. Similar difficulties are often encountered in shunted patients to predict the scope for improvement by shunt re-adjustment or revision. In this study we aimed to update our knowledge of how key quantitative parameters describing CSF dynamics may be used in diagnosis of shunt-responsive hydrocephalus and in the assessment of shunt function.

METHODS

A number of quantitative parameters [including resistance to CSF outflow (Rcsf), pulse amplitude of intracranial pressure waveform (AMP), RAP index and slow vasogenic waves] were studies in 1423 patients with 2665 CSF infusion tests and 305 overnight intracranial pressure (ICP)-monitoring sessions over a 17 year period.

OBSERVATIONS

We demonstrate our observations for typical values of Pb, Rcsf, AMP, slow vasogenic waves derived from infusion studies or overnight ICP monitoring in differentiating atrophy from shunt-responsive normal pressure hydrocephalus or acute hydrocephalus. From the same variables tested on shunted patients we demonstrate a standardised approach to help differentiate a properly-functioning shunt from underdrainage or overdrainage.

CONCLUSIONS

Quantitative variables derived from CSF dynamics allow differentiation between clinically overlapping entities such as shunt-responsive normal pressure hydrocephalus and brain atrophy (not shunt responsive) as well as allowing the detection of shunt malfunction (partial or complete blockage) or overdrainage. This observational study is intended to serve as an update for our understanding of quantitative testing of CSF dynamics.

A study of perioperative lumbar cerebrospinal fluid pressure in patients undergoing acoustic neuroma surgery

The objective of this study was to measure changes in cerebrospinal fluid (CSF) pressure and cerebrovascular hemodynamics following acoustic neuroma surgery. The subjects were 32 patients undergoing translabyrinthine or retrosigmoid excision of acoustic neuroma. CSF pressure and the amplitude of the CSF pressure pulse wave were measured using lumbar catheters, and all variables were recorded minute by minute on a microcomputer. Transcranial doppler (TCD) was used to measure flow velocity in the middle cerebral artery in 10 patients to monitor changes in cerebral hemodynamics. In the 24 hours after surgery, all patients showed a statistically significant rise in CSF pressure from 11.4 mm Hg (standard deviation [SD] 6.1) to 19.6 mm Hg (SD 5.2) and a corresponding fall in the compliance of the CSF compartment. These changes were reversed within 48 hours, and the CSF pressure fell below the preoperative level over the next 4 days without any drainage of CSF. The results of this study demonstrate a transient increase in CSF pressure and decrease in craniospinal compliance that is provoked by surgery. The most plausible explanation for this disturbance is impaired CSF absorption, which resolves rapidly in most patients without therapeutic CSF drainage.

Impact of duration of symptoms on CSF dynamics in idiopathic normal pressure hydrocephalus

OBJECTIVE

Cerebrospinal fluid (CSF) pressure-volume compensation may change over time as part of normal ageing, where the resistance to CSF outflow increases and the formation of CSF decreases with age. Is CSF compensation dependent on duration of symptoms in idiopathic normal pressure hydrocephalus (iNPH)?

METHODS

We investigated 92 patients presenting with iNPH. Mean age was 73 (range 47-86). There were 60 men and 32 women. They all presented with gait disturbance and ventricular dilatation. Memory deficit occurred in 72% and urinary incontinence in 52% of patients. All patients underwent computerized CSF infusion tests. Sixty-four shunted patients were available for follow-up, and their improvement was expressed using the NPH score.

RESULTS

Mean intracranial pressure (ICP) was 10.1±5.1 mmHg, and mean resistance to CSF outflow was 17.3±5.2 mmHg/(ml/min). Mean duration of symptoms was 24±19 months (range from 2 weeks to 86 months). Baseline ICP, magnitude of ICP pulse waveform, brain compliance and improvement after shunting (72% of patients improved) did not exhibit any dependency on the duration of symptoms. The resistance to CSF outflow showed a strong tendency to decrease in time with the duration of symptoms beyond 2 years (R= -0.702; P<0.005).

CONCLUSION

This is a preliminary observation, and it suggests that for patients with duration of symptoms longer than 2-3 years, the threshold for normal resistance to CSF outflow should be duration-adjusted.

Autonomic neuropathy is associated with impairment of dynamic cerebral autoregulation in type 1 diabetes

HYPOTHESIS

The mechanisms underlying impairment of dynamic cerebral autoregulation in diabetes are not well known. Cardiovascular autonomic neuropathy (CAN) could contribute to dynamic cerebral autoregulation impairment. In this study, we assessed the association between CAN and impairment of dynamic cerebral autoregulation in patients with type 1 diabetes.

METHODS

We evaluated dynamic cerebral autoregulation (DCA) in patients with type 1 diabetes and no history of cerebrovascular disease. DCA was assessed with transcranial Doppler using the correlation coefficient index Mx method. Mx was calculated from slow changes in mean cerebral blood flow velocity and mean arterial blood pressure. Increase in Mx indicates weaker DCA, with a threshold for impaired DCA above 0.3. Moderate CAN was defined as reduced heart rate variability (HRV) on the following tests: deep controlled breathing, Valsalva maneuver or initiation of active standing. Severe CAN was defined as reduced HRV associated with orthostatic hypotension.

RESULTS

60 patients were included (M/F: 33/27; mean age ± SD: 46 years ± 11.5). 23 patients had moderate CAN and 15 patients severe CAN. DCA was impaired in 37 patients. CAN was associated with impaired DCA (p = 0.005). Impairment of DCA was more pronounced in patients with severe CAN (p = 0.019). Glycosylated haemoglobin (HbA1c) was associated with impaired DCA in univariate analysis (p = 0.05). In multivariate analysis, only CAN was associated with impaired DCA (p = 0.007) whereas HbA1c was not (p = 0.161).

CONCLUSIONS

CAN was associated with impaired DCA in type 1 diabetes. The magnitude of DCA impairment increased with the severity of CAN.

Evaluation of the cerebrovascular pressure reactivity index using non-invasive finapres arterial blood pressure

A pressure reactivity index (PRx) can be assessed in patients with continuous monitoring of arterial blood pressure (ABP) and intracranial pressure (ICP) as a moving correlation coefficient between slow fluctuations of these two signals within a low frequency bandwidth. The study aimed to investigate whether the invasive ABP monitoring can be replaced with non-invasive measurement of ABP using a Finapres plethysmograph (fABP) to calculate the fPRx. There is a well-defined group of patients, suffering from hydrocephalus and undergoing CSF pressure monitoring, which may benefit from such a measurement. 41 simultaneous day-by-day monitoring of ICP, ABP and fABP were performed for about 30 min in 10 head injury patients. A Bland-Altman assessment for agreement was used to compare PRx and fPRx calculations. Performance metrics and the McNemary test were used to determine whether fPRx is sensitive enough to distinguish between functioning and disturbed cerebrovascular pressure reactivity. The fPRx correlated with PRx (R(Spearman) = 0.92, p < 0.001; bias = -0.04; lower and upper limits of agreement: -0.26 and 0.17, respectively). The fPRx distinguished between active and passive reactivity in more than 89% cases. The fPRx can be used with care for assessment of cerebrovascular reactivity in patients for whom invasive ABP measurement is not feasible. The fPRx is sensitive enough to distinguish between functional and deranged reactivity.

In vivo assessment of hydrocephalus shunt

OBJECTIVES

Over a 3-year period, we have performed 312 tests in 197 shunted patients. The data have been analyzed retrospectively to: (1) investigate the parameters describing CSF dynamics that correlate with shunt under-drainage and (2) estimate accuracy of this method.

METHODS

Constant rate infusion tests into shunt prechamber were performed.

RESULTS

In 161 of the 312 infusion tests, results indicated under-draining shunts. Patients in the under-draining group had higher baseline and plateau CSF pressures, higher resistance to CSF outflow and higher levels of baseline pulse amplitude waveform. During the test, a significantly greater vasogenic waves and lower compensatory reserve was noticed in patients with blocked shunts. In 21 patients with suggestion of shunt blockage and who subsequently underwent operative revision of the shunt, reports of intraoperative shunt patency were available. Shunt blockage was confirmed intra-operatively during surgery in 19 cases.

CONCLUSIONS

In vivo shunt testing is easy, safe and clinically useful, aiding decision in difficult clinical situations, where shunt malfunction is suspected but not certain. It also has satisfactory positive predictive power.

Link between vasogenic waves of intracranial pressure and cerebrospinal fluid outflow resistance in normal pressure hydrocephalus

Recent studies on normal pressure hydrocephalus (NPH) have pointed to a possible link between the disturbance in CSF circulation and cerebrovascular factors. We investigated the quantitative relationship between the resistance to CSF outflow (Rcsf) and vasogenic waves of ICP in patients with normal pressure hydrocephalus. Forty-five patients with NPH were investigated by an infusion study. The magnitudes of vasogenic ICP components: pulse, respiratory and slow vasogenic waves were assessed, and compared with Rcsf. Both baseline respiratory and slow waves of ICP were positively correlated with Rcsf. The respiratory wave at baseline was a single independent predictor of Rcsf (r = 0.66, p < 0.0002). All vasogenic components increased significantly during the infusion test. The magnitude of the increase was positively correlated with Rcsf. The vasogenic ICP waves, notably the respiratory wave of ICP, correlate with the resistance to CSF outflow.

How does CSF dynamics change after shunting?

OBJECTIVE

Hydrocephalus is much more complex than a simple disorder of cerebrospinal fluid (CSF) circulation. Shunting primarily corrects disturbed fluid flow which may have an impact on cerebral blood flow and metabolism. We studied hydrocephalic patients before and after shunting to characterize changes in their CSF compensatory parameters.

MATERIAL AND METHODS

We selected 25 patients and studied them retrospectively. All patients had ventriculomegaly and clinical symptoms of normal pressure hydrocephalus. After shunting, they were still presenting with some adverse symptoms, mainly headaches, slow improvement or no improvement of ventriculomegaly. Therefore, they underwent further infusion studies to assess shunt function. In all cases, the shunts were confirmed to be draining CSF adequately. Parameters of CSF dynamics: baseline intracranial pressure (ICP), resistance to CSF outflow, cerebrospinal elasticity, content of vasogenic pressure waves (pulse, respiratory and B waves) and compensatory reserve assessed as moving correlation coefficient between mean CSF pressure and pulse amplitude (RAP), were compared before and after shunting.

RESULTS

Mean ICP and resistance to CSF outflow decreased (P < 0.003) after shunting. All vasogenic pressure waves decreased (P < 0.005). Compensatory reserve (RAP) significantly improved (P < 0.005).

CONCLUSION

A functioning shunt has an important impact on CSF circulation and pressure-volume compensation. Infusion studies can demonstrate the return of disturbed CSF dynamics to normal values even if clinical or radiological changes are not dramatic.

Association between intracranial, arterial pulse pressure amplitudes and cerebral autoregulation in head injury patients

OBJECTIVE

To explore whether intracranial pulse pressure amplitudes relate to arterial pulse pressure amplitudes and whether correlations between time-related changes in intracranial and arterial pulse pressure amplitudes associate with indices of cerebral autoregulation.

METHODS

A total of 257 continuous and simultaneous intracranial pressure (ICP), arterial blood pressure (ABP) and middle cerebral artery (MCA) blood velocity recordings were obtained 1-14 days after ictus in 76 traumatic head injury patients and analysed retrospectively. Clinical outcome was assessed using the Glasgow outcome scale (GOS). Pulse pressure amplitudes of corresponding single ICP and ABP waves were correlated in consecutive 200 wave pairs. Mean ICP, mean ABP and mean ICP wave amplitudes, and mean and systolic MCA blood flow velocities, were computed in consecutive 6 second time windows. The indices of cerebral autoregulation PRx (moving correlation between mean ICP and mean ABP), and Mx and Sx (moving correlation between mean and systolic MCA blood velocity and cerebral perfusion pressure) were calculated over 4 minute periods and averaged over each recording.

RESULTS

Intracranial pulse pressure amplitudes were not related to arterial pulse pressure amplitudes (mean of Pearson's correlations coefficients: 0.04). Outcome was related to mean ICP, PRx and Sx (p </= 0.04, multiple regression analysis). Correlations between intracranial and arterial pulse pressure amplitudes were weakly related to PRx (Pearson's correlation coefficient: 0.16; p=0.01), but were not related to the indices of cerebral autoregulation Mx (Pearson's correlation coefficient: 0.07) and Sx (Pearson's correlation coefficient: 0.04).

CONCLUSIONS

In this cohort of pressure recordings, we found no evidence of a correlation between intracranial and arterial blood pressure amplitudes. The correlation appeared not to be related to the state of cerebral autoregulation, although a weak correlation was found with pressure reactivity index PRx.

Accuracy of non-invasive ICP assessment can be increased by an initial individual calibration

OBJECTIVE

In a formerly introduced mathematical model, intracranial pressure (ICP) could be non-invasively assessed using cerebral blood flow velocity (FV) and arterial blood pressure (ABP). The current study attempts to check whether the accuracy of the non-invasive ICP assessment (nICP) improves after an initial individual calibration by implanted ICP probes.

METHODS

Thirteen patients with brain lesions (35-77 years, mean: 58 +/- 13 years) were studied. FV, ABP and ICP signals were recorded at days 1, 2, 4 and 7. nICP was calculated and compared to ICP. In the first recording of each patient the (invasively assessed) ICP signal was used to calibrate the nICP calculation procedure, while the follow-up recordings were used for its validation.

FINDINGS

In 11 patients 22 follow-up recordings were performed. The mean deviation between ICP and the original nICP (+/- SD) was 8.3 +/- 7.9 mmHg. Using the calibrated method this deviation was reduced to 6.7 +/- 6.7 mmHg (P < 0.005).

CONCLUSIONS

Initial individual calibration of nICP assessment method significantly improves the accuracy of nICP estimation on subsequent days. This hybrid method of ICP assessment may be used in intensive care units in patients with initially implanted ICP probes. After removal of the probes, ICP monitoring can be continued using the calibrated nICP assessment procedure.

Ventriculostomy for control of raised ICP in acute traumatic brain injury

The aim of this study was to evaluate the effect of ventriculostomy on intracranial pressure (ICP), and related parameters, including cerebrospinal compensation, cerebral oxygenation (PbtO2) and metabolism (microdialysis) in patients with traumatic brain injury (TBI).

MATERIALS AND METHODS

Twenty-four patients with parenchymal ICP sensors were prospectively included in the study. Ventriculostomy was performed after failure to control ICP with initial measures. Monitoring parameters were digitally recorded before and after ventriculostomy and compared using appropriate tests.

RESULTS

In all patients ventriculostomy led to rapid reduction in ICP. Pooled mean daily values of ICP remained < 20mmHg for 72h after ventriculostomy and were lower than before (p < 0.001). In 11 out of 24 patients during the initial 24-h period following ventriculostomy an increase in ICP to values exceeding 20mmHg was observed. In the remaining 13 patients ICP remained stable, allowing reduction in the intensity of treatment. In this group ventriculostomy led to significant improvement in craniospinal compensation (RAP index), cerebral perfusion pressure and PbtO2. Improvement in lactate/pyruvate ratio, a marker of energy metabolism, was correlated with the increase in PbtO2.

CONCLUSION

Ventriculostomy is a useful ICP-lowering manoeuvre, with sustained ICP reduction and related physiological improvements achieved in > 50% of patients.

Pulse amplitude of intracranial pressure waveform in hydrocephalus

BACKGROUND

There is increasing interest in evaluation of the pulse amplitude of intracranial pressure (AMP) in explaining dynamic aspects of hydrocephalus. We reviewed a large number of ICP recordings in a group of hydrocephalic patients to assess utility of AMP.

MATERIALS AND METHODS

From a database including approximately 2,100 cases of infusion studies (either lumbar or intraventricular) and overnight ICP monitoring in patients suffering from hydrocephalus of various types (both communicating and non-communicating), etiology and stage of management (non-shunted or shunted) pressure recordings were evaluated. For subgroup analysis we selected 60 patients with idiopathic NPH with full follow-up after shunting. In 29 patients we compared pulse amplitude during an infusion study performed before and after shunting with a properly functioning shunt. Amplitude was calculated from ICP waveforms using spectral analysis methodology.

FINDINGS

A large amplitude was associated with good outcome after shunting (positive predictive value of clinical improvement for AMP above 2.5 mmHg was 95%). However, low amplitude did not predict poor outcome (for AMP below 2.5 mmHg 52% of patients improved). Correlations of AMP with ICP and Rcsf were positive and statistically significant (N = 131 with idiopathic NPH; R = 0.21 for correlation with mean ICP and 0.22 with Rcsf; p< 0.01). Correlation with the brain elastance coefficient (or PVI) was not significant. There was also no significant correlation between pulse amplitude and width of the ventricles. The pulse amplitude decreased (p < 0.005) after shunting.

CONCLUSIONS

Interpretation of the ICP pulse waveform may be clinically useful in patients suffering from hydrocephalus. Elevated amplitude seems to be a positive predictor for clinical improvement after shunting. A properly functioning shunt reduces the pulse amplitude.

ICM+, a flexible platform for investigations of cerebrospinal dynamics in clinical practice

BACKGROUND

ICM+ software encapsulates 20 years of our experience in brain monitoring gained in multiple neurosurgical and intensive care centres. It collects data from a variety of bedside monitors and produces on-line time trends of parameters defined using configurable signal processing formulas. The resulting data can be displayed in a variety of ways including time trends, histograms, cross histograms, correlations, etc. For technically minded researchers there is a plug-in mechanism facilitating registration of third party libraries of functions and analysis tools.

METHODS

The latest version of the ICM+ software has been used in 162 severely head injured patients in the Neurosciences Critical Care Unit of the Addenbrooke's Cambridge University Hospital. Intracranial pressure (ICP) and invasive arterial blood pressure (ABP) were monitored routinely. Mean values of ICP, ABP, cerebral perfusion pressure (CPP) and various indices describing pressure reactivity (PRx), pressure-volume compensation (RAP) and vascular waveforms of ICP were calculated. Error-bar chart showing reactivity index PRx versus CPP ('Optimal CPP' chart) was calculated continuously.

FINDINGS

PRx showed a significant relationship with CPP (ANOVA: p < 0.021) indicating loss of cerebral pressure-reactivity for low CPP (CPP < 55 mmHg) and for high CPPs (CPP > 95 mmHg). Examining PRx-CPP curves in individual patients revealed that CPP(OPT) not only varied between subjects but tended to fluctuate as the patient's state changed during the stay in the ICU. Calculation window of 6-8 h provided enough data to capture the CPP(OPT) curve.

CONCLUSIONS

ICM+ software proved to be useful both academically and clinically. The complexity of data analysis is hidden inside loadable profiles thus allowing clinically minded investigators to take full advantage of signal processing engine in their research into cerebral blood and fluid dynamics.