Effects of poor bone window on the assessment of cerebral autoregulation with transcranial Doppler sonography - a source of systematic bias and strategies to avoid it.

BACKGROUND: The consequences of poor insonation conditions on autoregulation parameters assessed with transcranial Doppler (TCD) are unclear. METHODS: We present two new complementary methods to assess the quality of a TCD signal. Inserting a thin aluminium foil between TCD probe and skin makes a simple model to artificially worsen a good insonation window. Validation studies are presented. We assessed insonation quality and cerebral autoregulation parameters with transfer function analysis and cross correlation in 46 healthy volunteers with and without the aluminium foil model. The same studies were operated on 45 patients with good insonation windows, naïve, after worsening the bone window and during constant infusion of an ultrasound contrast agent. For studying reproducibility, we assessed autoregulation twice in 30 patients with poor bone windows, with and without constant contrast infusion. RESULTS: Both methods to measure insonation quality are valid and reproducible. The aluminium foil model realistically simulates a natural poor bone window, reducing the signal quality (e.g. energy of the signal spectrum from 33.4+/-3.5 to 26.2+/-2.5 dB, p<0.001). Thereby, the autoregulation parameters are systematically biased (e.g. phase difference from 37.3+/-10.1 degrees to 25.9+/-15.1 degrees , p<0.001); while with the use of an ultrasound contrast agent this can be largely compensated (phase difference 35.7+/-10.7 degrees , p<0.001). The reproducibility is significantly improved (ICC from 0.76 to 0.90, p<0.05). CONCLUSIONS: Poor bone windows can cause considerable bias in TCD autoregulation parameters. This bias might be avoided by the use of ultrasound contrast agents, which may greatly improve the credibility of TCD autoregulation assessment in elderly patients.

Assessment of cerebral autoregulation with transcranial Doppler sonography in poor bone windows using constant infusion of an ultrasound contrast agent.

Cerebral autoregulation is an important pathophysiological and prognostic parameter for a variety of neurologic conditions. It can be assessed quickly and safely using transcranial Doppler sonography (TCD). In elderly patients, poor insonation conditions decrease the number of examinable patients and can cause a systematic bias in autoregulation parameters. The aim of this study was to investigate whether a constant infusion of an ultrasound contrast agent (Levovist((R))) can counteract these effects. We examined two cohorts of unselected neurologic patients. In 45 patients with good insonation windows (cohort 1), we used a thin aluminium foil between the skin and the TCD probe to artificially decrease the insonation quality. We determined two parameters of cerebral autoregulation (phase difference [PD] and a cross-correlation coefficient [Mx]) in native patients, with aluminium foil and with aluminium foil and a constant infusion of Levovist. In 30 patients with poor insonation windows (cohort 2), we measured the autoregulation twice, with and without an infusion of Levovist, to assess the reproducibility of the autoregulation parameters. In cohort 1, the foil model significantly decreased the Doppler signal quality, i.e., the mean spectrum energy decreased from 33.9 +/- 2.7 dB to 26.3 +/- 2.4 dB (p < 0.001). This introduced a significant bias to all autoregulation parameters (PD: decreased from 38.2 +/- 10.0 degrees to 27.9 +/- 12.5 degrees (p < 0.001); Mx: decreased from 0.308 +/- 0.170 to 0.254 +/- 0.162 (p < 0.01)). Both effects were compensated largely by a constant infusion of Levovist (300 mg/min). In cohort 2, infusion of the contrast agent at the same rate increased insonation quality, too, but to a lesser degree (27.4 +/- 2.4 dB to 32.0 +/- 3.7 dB, p < 0.001). This smaller increase did not cause a significant change in the autoregulation parameters, but the reproducibility of the PD was significantly improved (intraclass coefficient coefficient [ICC] 0.76, 95% confidence interval [0.59-0.87] in native poor bone window compared with ICC 0.90, 95% confidence interval [0.81-0.95] with infusion of the contrast agent). Our data show that constant infusion of an ultrasound contrast agent during the assessment of cerebral autoregulation can avoid potential bias introduced by poor insonation conditions. Furthermore, infusion of the contrast agent can improve reproducibility and contribute to the credibility of autoregulation assessment in the elderly. (E-mail: matthias.lorenz@em.uni-frankfurt.de).

Influence of temporal insonation window quality on the assessment of cerebral autoregulation with transcranial Doppler sonography.

Cerebral autoregulation can be assessed quickly and noninvasively using transcranial Doppler sonography (TCD). A poor transtemporal insonation window is a common limitation in TCD examinations. The effects of insonation quality on TCD autoregulation measurements have never been investigated. We developed two methods to quantitatively measure insonation quality. We also validated a model to artificially worsen the insonation window. This involves inserting a thin aluminium foil between the TCD probe and the skin. Forty-six healthy volunteers underwent TCD examination before and after artificial reduction of the transtemporal insonation quality. Two autoregulation parameters, the phase difference (PD) and a cross-correlation coefficient (Mx), were calculated from the recordings. The aluminium foil model realistically simulates poor insonation conditions, producing a decrease in the signal power from 36.4 +/- 2.8 dB to 32.0 +/- 1.5 dB. In corrupted insonation windows, the PD is artificially low (native: 34.2 +/- 9.3 degrees , corrupted: 29.9 +/- 9.7 degrees , p = 0.002), but the Mx is not significantly different. Insonation quality may introduce a systematic bias of clinically relevant magnitude into TCD-based cerebral autoregulation studies. This must be considered when studies of elderly patients are planned. This article discusses strategies to account for this bias.