Search Results

You are looking at 1 - 10 of 11 items for

  • Author or Editor: Piotr Smielewski x
  • All content x
Clear All Modify Search
Restricted access

Carole L. Turner, Susan Tebbs, Piotr Smielewski, and Peter J. Kirkpatrick

Object. Applanation tonometry is a noninvasive method of assessing both peripheral and central arterial blood pressure (BP) profiles. In this study the authors examine whether there are differences in these profiles in patients with intracranial aneurysms when compared with age-matched controls.

Methods. Carotid artery (CA) and derived aortic BP waveforms were obtained using a pulse wave analysis system. The ratio of the pressure wave amplitude above the systolic shoulder to the total systolic BP (augmentation index [AI]) was recorded.

One hundred seventy-three patients with intracranial aneurysms (23 unruptured lesions) and 173 healthy control volunteers were examined. For the patients with aneurysms the right and left CA AIs (mean ± standard deviation) were 125.6 ± 23.1% and 128.3 ± 22.1%, respectively. Corresponding values for the control group were 118.4 ± 22.6% and 119.4 ± 21.8%. The calculated AI for the ascending aorta was 29.8 ± 10.5% and 25.6 ± 12.2% for patients with aneurysms and control volunteers, respectively. Significant asymmetry in CA AI was seen in patients with aneurysms, the left being greater (p = 0.002). No significant differences were seen in mean BP (108 ± 14 mm Hg in patients with aneurysms compared with 106 ± 16 mm Hg in controls; p = 0.2). Multivariate analysis excluded the influence of BP and other potential confounding vascular risk factors for increased AI.

Conclusions. Significant differences in AI, both in magnitude and symmetry, were identified in patients with intracranial aneurysms when compared with matched controls.

Restricted access

Marek Czosnyka, Piotr Smielewski, Stefan Piechnik, Luzius A. Steiner, and John D. Pickard

Object. The goal of this study was to examine the relationship between cerebral autoregulation, intracranial pressure (ICP), arterial blood pressure (ABP), and cerebral perfusion pressure (CPP) after head injury by using transcranial Doppler (TCD) ultrasonography.

Methods. Using ICP monitoring and TCD ultrasonography, the authors previously investigated whether the response of flow velocity (FV) in the middle cerebral artery to spontaneous variations in ABP or CPP provides reliable information about cerebral autoregulatory reserve. In the present study, this method was validated in 187 head-injured patients who were sedated and receiving mechanical ventilation. Waveforms of ICP, ABP, and FV were recorded over intervals lasting 20 to 120 minutes. Time-averaged mean FV and CPP were determined. The correlation coefficient index between FV and CPP (the mean index of autoregulation [Mx]) was calculated over 4-minute epochs and averaged for each investigation.

The distribution of averaged mean FV values converged with the shape of the autoregulatory curve, indicating lower (CPP < 55 mm Hg) and upper (CPP > 105 mm Hg) thresholds of autoregulation. The relationship between the Mx and either the CPP or ABP was depicted as a U-shaped curve. Autoregulation was disturbed in the presence of intracranial hypertension (ICP ≥ 25 mm Hg) and when mean ABP was too low (ABP < 75 mm Hg) or too high (ABP > 125 mm Hg). Disturbed autoregulation (p < 0.005) and higher ICP (p < 0.005) occurred more often in patients with unfavorable outcomes than in those with favorable outcomes.

Conclusions. Autoregulation not only is impaired when associated with a high ICP or low ABP, but it can also be disturbed by too high a CPP. The Mx can be used to guide intensive care therapy when CPP-oriented protocols are used.

Restricted access

Peter J. Kirkpatrick, Joseph Lam, Pippa Al-Rawi, Piotr Smielewski, and Marek Czosnyka

Object. Signal changes in adult extracranial tissues may have a profound effect on cerebral near-infrared spectroscopy (NIRS) measurements. During carotid surgery NIRS signals provide the opportunity to determine the relative contributions from the intra- and extracranial vascular territories, allowing for a more accurate quantification. In this study the authors applied multimodal monitoring methods to patients undergoing carotid endarterectomy and explored the hypothesis that NIRS can define thresholds for cerebral ischemia, provided extracranial NIRS signal changes are identified and removed. Relative criteria for intraoperative severe cerebral ischemia (SCI) were applied to 103 patients undergoing carotid endarterectomy.

Methods. One hundred three patients underwent carotid endarterectomy. An intraoperative fall in transcranial Doppler—detected middle cerebral artery flow velocity (%ΔFV) of greater than 60% accompanied by a sustained fall in cortical electrical activity were adopted as criteria for SCI. Ipsilateral frontal NIRS recorded the total difference in concentrations of oxyhemoglobin and deoxyhemoglobin (Total ΔHbdiff). Interrupted time series analysis following clamping of the external carotid artery (ECA) and the internal carotid artery (ICA) allowed the different vascular components of Total ΔHbdiff (ECA ΔHbdiff and ICA ΔHbdiff) to be identified.

Data obtained in 76 patients were deemed suitable. A good correlation between %ΔFV and ICA ΔHbdiff (r = 0.73, p < 0.0001) was evident. Sixteen patients (21%) fulfilled the criteria for SCI. All patients who demonstrated an ICA ΔHbdiff of greater than 6.8 µmol/L showed SCI, and in two patients within this group nondisabling watershed infarction developed, as seen on postoperative computerized tomography scans. No patient with an ICA ΔHbdiff less than 5 µmol/L exhibited SCI or suffered a stroke. Within the resolution of the criteria used an ICA ΔHbdiff threshold of 6.8 µmol/L provided 100% specificity for SCI, whereas an ICA ΔHbdiff less than 5 µmol/L was 100% sensitive for excluding SCI. When Total ΔHbdiff was used without removing the ECA component, no thresholds for SCI were apparent.

Conclusions. Carotid endarterectomy provides a stable environment for exploring NIRS-quantified thresholds for SCI in the adult head.

Restricted access

Piotr Smielewski, Marek Czosnyka, Peter Kirkpatrick, and John D. Pickard

✓ The transient hyperemic response test has been shown to provide an index of cerebral autoregulation in healthy individuals and in patients who have suffered a subarachnoid hemorrhage. In this study, the test was applied to patients who had received a severe head injury, and the value of the test was assessed by comparing its result with the individual's clinical condition (Glasgow Coma Scale [GCS] score), cerebral perfusion pressure (CPP), transcranial Doppler wave form—derived index for cerebral autoregulation (relationship between the CPP and the middle cerebral artery flow velocity), and outcome (Glasgow Outcome Scale [GOS] score).

Forty-seven patients, aged 16 to 63 years, with head injuries were included in the study. Signals of intracranial pressure, arterial blood pressure, flow velocity, and cortical microcirculatory flux were digitized and recorded for a period of 30 minutes using special computer software. Two carotid compressions were performed at the beginning of each recording. The transient hyperemic response ratio (THRR: the ratio of the hyperemic flow velocity recorded after carotid release and the precompression baseline flow velocity) was calculated, as was the correlation coefficient Sx used to describe the relationship between slow fluctuations in the systolic flow velocity and CPP throughout the period of recording.

No significant changes in CPP were found during compression. There was a significant correlation between the THRR and the Sx (r = 0.49, p < 0.0001). The hyperemic response proved to be lower in patients who exhibited a poor clinical grade at presentation (GCS scores < 6, p = 0.01) and lower in patients achieving a poor outcome (GOS scores of 3, 4, and 5, p = 0.003). Loss of postcompression hyperemia occurred when the CPP fell below 50 mm Hg.

The carotid compression test provides a simple index of cerebral autoregulation that is relevant to the clinical condition and outcome of the severely head injured patient.

Restricted access

Marek Czosnyka, Marcella Balestreri, Luzius Steiner, Piotr Smielewski, Peter J. Hutchinson, Basil Matta, and John D. Pickard

Object. The object of this study was to investigate whether a failure of cerebrovascular autoregulation contributes to the relationship between age and outcome in patients following head injury.

Methods. Data obtained from continuous bedside monitoring of intracranial pressure (ICP), arterial blood pressure (ABP), and cerebral perfusion pressure (CPP = ABP — ICP) in 358 patients with head injuries and intermittent monitoring of transcranial Doppler blood flow velocity (FV) in the middle cerebral artery in 237 patients were analyzed retrospectively. Indices used to describe cerebral autoregulation and pressure reactivity were calculated as correlation coefficients between slow waves of systolic FV and CPP (autoregulation index [ARI]) and between ABP and ICP (pressure reactivity index [PRI]).

Older patients had worse outcomes after brain trauma than younger patients (p = 0.00001), despite the fact that the older patients had higher initial Glasgow Coma Scale scores (p = 0.006). When age was considered as an independent variable, it appeared that ICP decreased with age (p = 0.005), resulting in an increasing mean CPP (p = 0.0005). Blood FV was not dependent on age (p = 0.58). Indices of autoregulation and pressure reactivity demonstrated a deterioration in cerebrovascular control with advancing age (PRI: p = 0.003; ARI: p = 0.007).

Conclusions. An age-related decline in cerebrovascular autoregulation was associated with a relative deterioration in outcome in elderly patients following head trauma.

Restricted access

Marek Czosnyka, Basil F. Matta, Piotr Smielewski, Peter J. Kirkpatrick, and John D. Pickard

Object. The authors studied the reliability of a new method for noninvasive assessment of cerebral perfusion pressure (CPP) in head-injured patients in which mean arterial blood pressure (ABP) and transcranial Doppler middle cerebral artery mean and diastolic flow velocities are measured.

Methods. Cerebral perfusion pressure was estimated (eCPP) over periods of continuous monitoring (20 minutes—2 hours, 421 daily examinations) in 96 head-injured patients (Glasgow Coma Scale score < 13) who were admitted to the intensive care unit. All patients were sedated, paralyzed, and ventilated. The eCPP and the measured CPP (ABP minus intracranial pressure, measured using an intraparenchymal microsensor) were compared.

The correlation between eCPP and measured CPP was r = 0.73; p < 10−6. In 71% of the examinations, the estimation error was less than 10 mm Hg and in 84% of the examinations, the error was less than 15 mm Hg. The method had a high positive predictive power (94%) for detecting low CPP (< 60 mm Hg). The eCPP also accurately reflected changes in measured CPP over time (r > 0.8; p < 0.001) in situations such as plateau and B waves of intracranial pressure, arterial hypotension, and refractory intracranial hypertension. A good correlation was found between the average measured CPP and eCPP when day-by-day variability was assessed in a group of 41 patients (r = 0.71).

Conclusions. Noninvasive estimation of CPP by using transcranial Doppler ultrasonography may be of value in situations in which monitoring relative changes in CPP is required without invasive measurement of intracranial pressure.

Restricted access

Peter J. Kirkpatrick, Piotr Smielewski, Marek Czosnyka, David K. Menon, and John D. Pickard

✓ A multimodality recording system was used in 14 ventilated patients with closed head injury to assess the potential use of near-infrared spectroscopy (NIRS) in the neurointensive care unit. Signals of intracranial pressure, cerebral perfusion pressure, peripheral oxygen saturation, jugular venous saturation, and NIRS-derived changes in the chromophores of oxy- and deoxyhemoglobin were digitized and recorded. After a review of 886 hours of continuous monitoring, 376 hours were considered free from artifact and were entered for final analysis. In nine of the patients 38 events were recorded that demonstrated clear changes in cerebral perfusion pressure accompanied by hemodynamic changes in middle cerebral artery flow velocity (transcranial Doppler) and cortical perfusion (laser Doppler flowmetry). Near-infrared spectroscopy showed correlated changes in 37 events (97%) whereas jugular venous saturation monitoring registered only 20 (53%). There was associated peripheral oxygen desaturation in eight cases (21%), intracranial hypertension in 10 (26%), and cerebral hyperemia in eight (21%). The remaining 12 events (32%) appeared to be complex changes of uncertain origin. Iatrogenic factors were identified as causative in 14 cases (37%). The potential application of NIRS in adults and the importance of using multiple parameter recording systems in the interpretation of cerebral events are discussed.

Restricted access

Eric A. Schmidt, Marek Czosnyka, Luzius A. Steiner, Marcella Balestreri, Piotr Smielewski, Stefan K. Piechnik, Basil F. Matta, and John D. Pickard

Object. The aim of this study was to assess the asymmetry of autoregulation between the left and right sides of the brain by using bilateral transcranial Doppler ultrasonography in a cohort of patients with head injuries.

Methods. Ninety-six patients with head injuries comprised the study population. All significant intracranial mass lesions were promptly removed. The patients were given medications to induce sedation and paralysis, and artificial ventilation. Arterial blood pressure (ABP) and intracranial pressure (ICP) were monitored in an invasive manner. A strategy based on the patient's cerebral perfusion pressure (CPP = ABP − ICP) was applied: CPP was maintained at a level higher than 70 mm Hg and ICP at a level lower than 25 mm Hg. The left and right middle cerebral arteries were insonated daily, and bilateral flow velocities (FVs) were recorded. The correlation coefficient between the CPP and FV, termed Mx, was calculated and time-averaged over each recording period on both sides. An Mx close to 1 signified that slow fluctuations in CPP produced synchronized slow changes in FV, indicating a defective autoregulation. An Mx close to 0 indicated preserved autoregulation. Computerized tomography scans in all patients were reviewed; the side on which the major brain lesion was located was noted and the extent of the midline shift was determined. Outcome was measured 6 months after discharge. The left—right difference in the Mx between the hemispheres was significantly higher in patients who died than in those who survived (0.16 ± 0.04 compared with 0.08 ± 0.01; p = 0.04). The left—right difference in the Mx was correlated with a midline shift (r = −0.42; p = 0.03). Autoregulation was worse on the side of the brain where the lesion was located (p < 0.035).

Conclusions. The left—right difference in autoregulation is significantly associated with a fatal outcome. Autoregulation in the brain is worse on the side ipsilateral to the lesion and on the side of expansion in cases in which there is a midline shift.

Full access

Marek Czosnyka, Zofia Czosnyka, Nicole Keong, Andreas Lavinio, Piotr Smielewski, Shahan Momjian, Eric A. Schmidt, Gianpaolo Petrella, Brian Owler, and John D. Pickard


Apart from its mean value, the pulse waveform of intracranial pressure (ICP) is an essential element of pressure recording. The authors reviewed their experience with the measurement and interpretation of ICP pulse amplitude by referring to a database of recordings in hydrocephalic patients.


The database contained computerized pressure recordings from 2100 infusion studies (either lumbar or intraventricular) or overnight ICP monitoring sessions in patients suffering from hydrocephalus of various types (both communicating and noncommunicating), origins, and stages of management (shunt or no shunt). Amplitude was calculated from ICP waveforms by using a spectral analysis methodology.


The appearance of a pulse waveform amplitude is positive evidence of a technically correct recording of ICP and helps to distinguish between postural and vasogenic variations in ICP. Pulse amplitude is significantly correlated with the amplitude of cerebral blood flow velocity (R = 0.4, p = 0.012) as assessed using Doppler ultrasonography. Amplitude is positively correlated with a mean ICP (R = 0.21 in idiopathic normal-pressure hydrocephalus [NPH]; number of cases 131; p < 0.01) and resistance to cerebrospinal fluid outflow (R = 0.22) but does not seem to be correlated with cerebrospinal elasticity, dilation of ventricles, or severity of hydrocephalus (NPH score). Amplitude increases slightly with age (R = 0.39, p < 0.01; number of cases 46). A positive association between pulse amplitude and increased ICP during an infusion study is helpful in distinguishing between hydrocephalus and predominant brain atrophy. A large amplitude is associated with a good outcome after shunting (positive predictive power 0.9), whereas a low amplitude has no predictive power in outcome prognostication (0.5). Pulse amplitude is reduced by a properly functioning shunt.


Proper recording, detection, and interpretation of ICP pulse waveforms provide clinically useful information about patients suffering from hydrocephalus.

Restricted access

Marek Czosnyka, Piotr Smielewski, Stefan Piechnik, Eric A. Schmidt, Pippa G. Al-Rawi, Peter J. Kirkpatrick, and John D. Pickard

Object. Plateau waves of intracranial pressure (ICP) are often recorded during intensive care monitoring of severely head injured patients. They are traditionally interpreted as meaningful secondary brain insults because of the dramatic decrease in cerebral perfusion pressure (CPP). The aim of this study was to investigate both the hemodynamic profile and the clinical consequences of plateau waves.

Methods. One hundred sixty head-injured patients were studied using continuous monitoring of ICP; almost 20% of these patients exhibited plateau waves. In 96 patients arterial pressure, ICP, and transcranial Doppler (TCD) blood flow velocity were studied daily for 20 minutes to 3 hours. Sixteen episodes of plateau waves in eight patients were recorded and analyzed.

The dramatic increase in ICP was followed by a profound fall in CPP (by 45%). In contrast, flow velocity fell by only 20%. Autoregulation was documented to be intact both before and after plateau but was disturbed during the wave (p < 0.05). Pressure-volume compensatory reserve was always depleted before the wave. Cerebrovascular resistance decreased during the wave by 60% (p <0.05) and TCD pulsatility increased (p <0.05). Plateau waves did not increase the probability of an unfavorable outcome following injury.

Conclusions. The authors have confirmed that the plateau waves are a hemodynamic phenomenon associated with cerebrovascular vasodilation. They are observed in patients with preserved cerebral autoregulation but reduced pressure-volume compensatory reserve.