Journal of the Neurological Sciences 350 (2015) 33–39

Contents lists available at ScienceDirect

Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns

Association between ventricular volume measures and pulsatile and static intracranial pressure scores in non-communicating hydrocephalus Terje Sæhle a, Per Kristian Eide a,b,⁎ a b

Department of Neurosurgery, Oslo University Hospital–Rikshospitalet, Oslo, Norway Faculty of Medicine, University of Oslo, Oslo, Norway

a r t i c l e

i n f o

Article history: Received 2 December 2014 Received in revised form 16 January 2015 Accepted 1 February 2015 Available online 7 February 2015 Keywords: Non-communicating hydrocephalus Ventricular volume Pulsatile ICP Static ICP Ventricular growth Hydrocephalus pathogenesis

a b s t r a c t Background: In non-communicating hydrocephalus (HC), enlarged cerebral ventricles are often thought to reflect increased intracranial pressure (ICP) or increased pulsatile ICP. The present study was undertaken to explore the association between ventricular volume measures and pulsatile or static ICP scores in patients with noncommunicating HC. Since linear measures of ventricular size have the most widespread use, we also examined how linear and volume measures of ventricular size compare. Methods: The patient material includes all patients with non-communicating HC that underwent continuous over-night ICP monitoring during the period 2002–2011. The scores of pulsatile and static ICP were determined from the continuous ICP signals stored on the hospital server. Ventricular volume was determined both as linear measures of sectional CT or MR images and as 3D volume of all ventricles. We also determined the ventricular volume index as a relationship between ventricular volume and intracranial volume. Results: Eighty-five patients were included in the study; they were dichotomized into those that previously had not received endoscopic third ventriculostomy (ETV; n = 52; Group 1), and those that had previously underwent ETV (n = 33; Group 2). None was previously shunted. We found no significant correlations between the ICP scores and the ventricular volume indices in neither of the patient groups. In Group 1, however, the mean ICP wave amplitude was significantly higher than in Group 2. There was a strong positive correlation between volume and linear measures of ventricular size. We found neither any association between age and ventricular volume; nor any association between ventricular volume and duration of symptoms. Conclusions: In this cohort of patients with non-communicating HC, we found no evidence of a proportional correlation between ventricular volume and pulsatile or static ICP. However, the findings suggest that symptomatic and untreated non-communication HC is still associated with reduced intracranial compliance. © 2015 Elsevier B.V. All rights reserved.

1. Introduction The major neurosurgical treatments of hydrocephalus (HC) include shunting of communicating HC and endoscopic third ventriculostomy (ETV) of non-communicating HC [1,2]. The aim of both treatment modalities is to reduce raised intracranial pressure (ICP). The indication for surgery in non-communicating HC is primarily based on the history, clinical symptoms and the degree of ventriculomegaly [3]. Enlarged ventricles are often thought to be indicative of increased ICP. Given the fact that modern HC treatment (shunting or ETV) usually aims at reducing ventricular size and ICP, it is important to establish information how ventricular size and ICP associate. Increased knowledge about the underlying mechanisms behind

⁎ Corresponding author at: Department of Neurosurgery, Oslo University Hospital– Rikshospitalet, PB 4950 Nydalen, 0424 Oslo, Norway. Tel.: + 47 23074300; fax: + 47 23074310. E-mail addresses: [email protected], [email protected] (P.K. Eide).

http://dx.doi.org/10.1016/j.jns.2015.02.003 0022-510X/© 2015 Elsevier B.V. All rights reserved.

current neurosurgical treatment modalities of HC is required to improve management. The association between ventriculomegaly and ICP has not been established. According to the conventional view, considering HC as a disturbance of the cerebrospinal fluid (CSF) bulk flow, it is assumed that blockade of CSF flow causes increased CSF pressure within the ventricles, thereby causing the ventricles to grow. In several currently acceptable pulsatile hydrodynamic theories, HC can be explained by an increase in ventricular pulsatile pressure [4,5]. These models are mostly used for communicating HC, but have also been used to describe cases of non-communicating HC [6]. It has been hypothesized that acute HC may be caused by an intraventricular CSF bulk flow obstruction, while the ventriculomegaly in chronic obstructive HC is maintained by decreased intracranial compliance causing increased capillary pulsations [5]. An association between increased pulsatile pressure and ventriculomegaly is also supported by the findings of several studies [6–9]. Hence, gradients in pulsatile ICP may be a possible cause of ventriculomegaly [7,9]. However, no proportional correlation between the pulsatile ICP or mean ICP and the degree of ventriculomegaly has been shown. Hence,

34

T. Sæhle, P.K. Eide / Journal of the Neurological Sciences 350 (2015) 33–39

in patients with an established adult communicating HC, we previously reported no ventricular-subdural transmantle gradient [10], though uneven distribution of pulsatile ICP can be seen under certain circumstances in HC patients [11]. With regard to static ICP, the existence of a transmantle pressure gradient was not found [12]. The present study was undertaken to examine whether there is an association between ventricular volume measures and scores of pulsatile and static ICP. The hypothesis was that increased degree of ventriculomegaly is accompanied by more abnormal ICP in accordance with conventional theories about HC, or by more abnormal ICP pulsations in accordance with several hydrodynamic theories. For years, we have used linear measures to assess ventricular size [13]. Therefore, in this study we also determined the association between volume and linear measures of ventricular size. 2. Methods 2.1. Patients and study design The patient material includes all patients with non-communicating HC undergoing diagnostic ICP-wave/ICP monitoring within the Department of Neurosurgery, Oslo University Hospital–Rikshospitalet, during the period 2002–2011. Patients with shunts were not included. The study was approved by the Oslo University Hospital– Rikshospitalet as a quality study (Approval 10/16550). The Regional Committee for Medical and Health Research Ethics (REK) of Health Region South-East, Norway was informed in writing, and had no objections to the study. 2.2. Radiological assessment of ventricular size Two different approaches were used to determine ventricular size: (1) Ventricular volume was calculated using the integrated 3D volume functions in iPlan® software (BrainLAB, Munich, Germany), and expressed in cm3. The measured volume included the 3-dimensional space of the lateral ventricles, 3rd ventricle, aqueduct and 4th ventricle (Fig. 1). In addition, the total intracranial volume, including both the supra- and infratentorial compartments, was determined. The ventricular volume index was calculated as a percentage of ventricular volume of total intracranial volume. (2) Linear measures of ventricular size were determined as previously described [14], based on sectional computer tomography (CT)

or magnetic resonance imaging (MRI) pictures. Based on these measures, we calculated the Evan's index, 3rd ventricle index, cella media index and the ventricular score.

2.3. Monitoring of pulsatile and static ICP The institutional routine for ICP monitoring has previously been described. The indication for ICP monitoring was to determine requirement for surgery (ETV or shunting). In short, under local anesthesia or general anesthesia, a burr hole was made in the skull. A Codman ICP micro-sensor (Codman Microsensor, Codman MicroSensor, Johnson and Johnson, Raynham, Massachusetts, USA) was zeroed against the atmospheric pressure, and placed 1–2 cm into the brain parenchyma via a minor opening in the dura. The patient was then returned to the ward and the ICP monitoring started and continued until the next day. The continuous ICP waveforms sampled at 100–200 Hz are stored in a hospital server. For this study, the continuous ICP waveforms of the included patients were retrieved from the database, and analyzed using a previously published method for automatic analysis of cardiac induced ICP waves (Sensometrics Software, dPCom As, Oslo) [15]. Each cardiacinduced wave is automatically identified by its beginning and ending diastolic minimum pressure, and by its systolic maximum pressure. From the identified cardiac-beat-induced ICP waves, the amplitude (‘pulse amplitude’ dP), the rise time (dT) and rise time coefficient (RTC, dP/dT) are automatically determined. Subsequently, for each 6-s time window, the ICP waveform parameters mean wave amplitude (MWA), mean wave rise time (RT) and mean wave rise time coefficient (RTC) are determined. Only 6-s time windows containing minimum four cardiac beat induced waves were considered to be of acceptable quality and used for the present analysis. The automatic method also identifies artifact waves due to noise in the pressure signal, and these were omitted from the analysis. Moreover, for each 6-s time window the conventional static ICP is determined as the mean ICP. In order to compare the pressure scores between patients, we used a standardized recording time from 23 p.m. until 7 a.m. This recording time was also compared with another recording time from 7 a.m. until 10 a.m. 3. Statistical analysis Statistical analysis was performed using SPSS Statistics version 20 (IBM Corporation, Armonk, NY). Differences between groups were determined using independent sample t-test. Associations between observations were determined by the Pearson correlation coefficient. Significance was accepted at the .05 level. 4. Results 4.1. Patients In total 86 patients with non-communicating HC underwent ICP monitoring during the 10-year period 2002–2011. Group 1 consisted of the 53 patients in whom no previous shunting or ETV had been done. Group 2 consisted of the remaining 33 patients who previously had undergone ETV. None had a shunt. The demographic data are presented in Table 1. 4.2. Ventricular volume and ICP scores

Fig. 1. Illustration of 3D volume measures in iPlan® software.

Both the ventricular volume, ventricular volume index, and the linear measures of ventricular size, were comparable in the two groups (Table 2). Except for a difference in cella media index, no significant differences between groups were seen.

T. Sæhle, P.K. Eide / Journal of the Neurological Sciences 350 (2015) 33–39 Table 1 Patient material.

N Age (years) Gender (F/M) Symptom duration (months)

Table 3 Scores of static and pulsatile ICP*. Group 1 (no previous ETV)

Group 2 (previous ETV)

52 50 (1–78)a 22/30 12 (0–120)

33 26 (4–68) 17/16 6 (2–60)

Independent sample t-test comparing Groups 1 and 2. a p b 0.01.

The average pulsatile and static ICP scores are presented in Table 3. The MWA values were significantly lower and the rise time (RT) was significantly shorter in Group 2 with previous ETV (Table 3). 4.3. Association between ICP waves and ventricular volume We found no significant association between the ventricular volume or ventricular volume index and the pulsatile (MWA, RT or RTC) or static (mean ICP) ICP scores in patients previously not treated with ETV (Fig. 2). Similar observations were done in those with previous ETV (Group 2, data not shown). 4.4. Association between volume and linear indices of ventricular size Both ventricular volume and ventricular volume index were significantly associated with the linear measures of ventricular size in Group 1 (Fig. 3). We made comparable observations in patients of Group 2 (data not shown).

There was no significant association between ventricular volume/ ventricular volume index and age or the duration of the symptoms (Fig. 4). 5. Discussion In the present cohorts of non-communicating HC, there was no proportional association between ventricular volume measures and scores of pulsatile or static ICP. Regarding various measures of ventricular size, we found a close association between 3D volume calculations and linear measures of ventricular size on sectional images of the brain. 5.1. Patient material The diagnosis of non-communicating HC was based on MRI and evidence of aqueductal stenosis. During this period phase-contrast MRI,

Table 2 Measures of ventricular size. Group 1 (no previous ETV)

Group 2 (previous ETV)

163 (29–780) 1450 (1023–2402) 12 (2–33)

165 (28–742) 1633 (1114–2116) 12 (2–36)

0.41 (0.25–0.61) 0.29 (0.14–0.49)a 0.11 (0.04–0.21) 105 (59–155)

0.42 (0.28–0.53) 0.36 (0.18–0.54) 0.11 (0.04–0.21) 115 (57–194)

Independent sample t-test comparing Groups 1 and 2. a p b 0.01.

ICP wave parameters Mean wave amplitude (MWA) Average (mm Hg) Percentage ≥5 mm Hg Percentage ≥6 mm Hg Mean wave rise time (RT) Average (s) Percentage ≥0.20 s Percentage ≥0.25 s Mean wave rise time coeff. (RTC) Average (mm Hg/s) Percentage ≥30 mm Hg/s Percentage ≥40 mm Hg/s Static ICP Mean ICP Average (mm Hg) Percentage ≥15 mm Hg Percentage ≥20 mm Hg

Group 1 (no previous ETV)

Group 2 (previous ETV)

4.4 (2.2–15.8)a 29 (0–100)a 10 (0–100)a

4.0 (2.5–9.1) 12 (0–89) 3 (0–71)

0.23 (0.09–0.31)b 89 (0–100)b 42 (0–100)b

0.15 (0.10–0.26) 16 (0–99) 6 (0–79)

23.0 (10.8–67.4) 14 (0–100) 2 (0–99)

28.6 (12.8–47.2) 47 (0–99) 4 (0–71)

9.9 (−9.2–49.1) 5 (0–100) 1 (0–100)

9.7 (−3.5–28.4) 2 (0–97) 0 (0–76)

*ICP parameters recorded from 11 pm to 7 am. ICP. Independent sample t-test comparing Groups 1 and 2. a p b 0.05. b p b 0.001.

which is most sensitive in diagnosing aqueductal stenosis, was not implemented on a regular basis. While Group 1 consisted of patients without any prior surgery, and can be regarded as true non-communicating HC, Group 2 had previously received EVT and can be considered as communicating HC. Notably, comparable results were seen in both groups. 5.2. Association between ventricular volume measures and static ICP scores

4.5. Association between ventricular volume and age

Volume measures Ventricular volume (ml) Intracranial volume (ml) Ventricular volume index Linear measures Evan's index Cella media index Third ventricle index Ventricular score

35

To our knowledge, the association between cardiac induced mean ICP scores and ventricular volume has previously not been determined in chronic non-communicating HC. The present data gave no support to the hypothesis that ventriculomegaly in non-communication HC is likely to be accompanied by elevated static ICP. Thus, the present observations also challenge the concept that enlarged ventricles reflect increased ICP in chronic noncommunicating HC. One alternative explanation to the present observations is that enlarged ventricles may be compensatory to increased ICP in some patients. It may also be counteracted by the corresponding changes in parenchymal pressure or compliance. In some studies of chronic HC, ventriculomegaly is found to be accompanied by increased ICP [16], but any proportional correlation between the degree of ventriculomegaly and ICP is not described. 5.3. Association between ventricular volume measures and pulsatile ICP scores The increased MWA in Group 1 compared with Group 2 supports the theory of improved intracranial compliance after treatment with EVT and the theories describing increased pulsation and impaired intracranial compliance in chronic HC [5]. Intracranial compliance refers to the pressure–volume reserve capacity of the intracranial compartment. Impaired intracranial compliance means that a small intracranial volume change results in a disproportional pressure increase, and elevated pulse pressure amplitudes [17]. The lack of a proportional relationship between the ventricular size and the ICP wave amplitude suggests a more complex model than the conventional theories about HC and the hydrodynamic theories of HC. Alternative explanations to this observation may be that increasing ventricular size is a compensatory event to impaired intracranial compliance or vice versa. According to pulsatile models, ventriculomegaly

36

T. Sæhle, P.K. Eide / Journal of the Neurological Sciences 350 (2015) 33–39

Fig. 2. Association between ventricular volume and (a) MWA, (b) RT, (c) RTC, and (d) mean ICP. Association between ventricular volume index and (e) MWA, (f) RT, (g) RTC, and (h) mean ICP. The Pearson correlation coefficient (R) and significance level are presented for each plot.

T. Sæhle, P.K. Eide / Journal of the Neurological Sciences 350 (2015) 33–39

37

Fig. 3. Association between ventricular volume and (a) Evans index, (b) third ventricular index, (c) cella media index, and (d) ventricular score. Association between ventricular volume index and (e) Evans index, (f) third ventricular index, (g) cella media index, and (h) ventricular score. The Pearson correlation coefficient (R) and significance level are presented for each plot.

38

T. Sæhle, P.K. Eide / Journal of the Neurological Sciences 350 (2015) 33–39

Fig. 4. Association between age and (a) ventricular volume and (b) ventricular volume index, and between symptom duration and (c) ventricular volume and (d) ventricular volume index. The Pearson correlation coefficient (R) and significance level are presented for each plot.

may be caused by increased ventricular pressure pulsations [6,7,9,18]. On the other hand, it can be hypothesized that the increased capillary pulsations in the cerebral parenchyma may further impair intracranial compliance and thereby counteract the growth of ventriculomegaly until a steady-state is reached. Individual differences in the degree of ventriculomegaly may depend on the balance between the pressure pulsations in the ventricles and the cerebral parenchyma. The ventricular size is also likely to be affected by multiple other factors, making the pathogenesis even more complex than shown in this model.

5.4. Volume versus linear measures of ventricular size Linear measures of ventricular size may be an inaccurate method due to differences in the positioning and orientation of the head during imaging or due to measurements at different sectional levels. However, the significance of this inaccuracy is unclear. The present data showed a good correlation between volume and linear measures of ventricular size. Thus, while linear measures provide a less accurate indication of ventricular size, it seems to provide a good approximation. Others have previously questioned that linear measures, such as Evan's index, accurately reflect ventricular volume [19]. In this study, the volume was calculated from the images taken in conjunction with the ICP monitoring. Although MRI was used in most of the cases, a CT scan was done in some of the patients. The differences in the imaging modalities

between different patients may have introduced a minor bias. However, the comparison of absolute ventricle volumes corresponded well with the comparison of ventricular volume in relation to intracranial volume, suggesting that calibration error of the images was not a significant bias. 6. Limitations Repeated measurements of pulsatile and static ICP during ventricular growth might show a closer proportional association between ventriculomegaly and pulsatile/static ICP scores than single measurements done after ventriculomegaly has been established. Moreover, pressure measurements within the brain parenchyma may not properly reflect gradients in pulsatile ICP between ventricular fluid, brain parenchyma, and subdural fluid. We have observed uneven distribution of pulsatile ICP under some circumstances in hydrocephalic patients [11]. Thus, the time of monitoring during development of hydrocephalus may be crucial. 7. Conclusions In the present cohorts of patients with non-communicating HC, we found no evidence of proportional correlation between ventricular volume measures and pulsatile/static ICP scores. The results question the concept that increased ventricular volume relates to increased ICP or

T. Sæhle, P.K. Eide / Journal of the Neurological Sciences 350 (2015) 33–39

increased pulsations in non-communicating HC. However, the findings support the concept of an impaired intracranial compliance in nontreated non-communication HC. Funding/financial support The study was supported by the Department of Neurosurgery, Oslo University Hospital–Rikshospitalet, Oslo, Norway. Contributions TS and PKE contributed to the conception of the study. TS made the bulk of the drafting of the article, and PKE contributed with thorough editing of the manuscript. Both authors have read and approved the final manuscript. Conflicts of interest Terje Sæhle MD discloses no conflicts of interest. Per Kristian Eide MD PhD has a financial interest in the software company (dPCom AS, Oslo) manufacturing the software (Sensometrics Software) used for the analysis of the ICP recordings. References [1] Edwards RJ, Dombrowski SM, Luciano MG, Pople IK. Chronic hydrocephalus in adults. Brain Pathol 2004;14:325–36. [2] Rasul FT, Marcus HJ, Toma AK, Thorne L, Watkins LD. Is endoscopic third ventriculostomy superior to shunts in patients with non-communicating hydrocephalus? A systematic review and meta-analysis of the evidence. Acta Neurochir (Wien) 2013;155:883–9. [3] Vogel TW, Bahuleyan B, Robinson S, Cohen AR. The role of endoscopic third ventriculostomy in the treatment of hydrocephalus. J Neurosurg Pediatr 2013;12: 54–61.

39

[4] Egnor M, Zheng L, Rosiello A, Gutman F, Davis R. A model of pulsations in communicating hydrocephalus. Pediatr Neurosurg 2002;36:281–303. [5] Greitz D. The hydrodynamic hypothesis versus the bulk flow hypothesis. Neurosurg Rev 2004;27:299–300. [6] Matsumoto T, Nagai H, Kasuga Y, Kamiya K. Changes in intracranial pressure (ICP) pulse wave following hydrocephalus. Acta Neurochir (Wien) 1986;82:50–6. [7] Di Rocco C, Pettorossi VE, Caldarelli M, Mancinelli R, Velardi F. Communicating hydrocephalus induced by mechanically increased amplitude of the intraventricular cerebrospinal fluid pressure: experimental studies. Exp Neurol 1978;59:40–52. [8] Wagshul ME, Eide PK, Madsen JR. The pulsating brain: a review of experimental and clinical studies of intracranial pulsatility. Fluids Barriers CNS 2011;8:5. [9] Penn RD, Basati S, Sweetman B, Guo X, Linninger A. Ventricle wall movements and cerebrospinal fluid flow in hydrocephalus. J Neurosurg 2011;115:159–64. [10] Eide PK, Saehle T. Is ventriculomegaly in idiopathic normal pressure hydrocephalus associated with a transmantle gradient in pulsatile intracranial pressure? Acta Neurochir (Wien) 2010;152:989–95. [11] Eide PK. Demonstration of uneven distribution of intracranial pulsatility in hydrocephalus patients. J Neurosurg 2008;109:912–7. [12] Stephensen H, Tisell M, Wikkelso C. There is no transmantle pressure gradient in communicating or noncommunicating hydrocephalus. Neurosurgery 2002;50: 763–71 [discussion 71–3]. [13] Eide PK. The relationship between intracranial pressure and size of cerebral ventricles assessed by computed tomography. Acta Neurochir (Wien) 2003;145:171–9 [discussion 9]. [14] Eide PK. Intracranial pressure parameters in idiopathic normal pressure hydrocephalus patients treated with ventriculo-peritoneal shunts. Acta Neurochir (Wien) 2006;148:21–9 [discussion 9]. [15] Eide PK. A new method for processing of continuous intracranial pressure signals. Med Eng Phys 2006;28:579–87. [16] Oi S, Shimoda M, Shibata M, Honda Y, Togo K, Shinoda M, et al. Pathophysiology of long-standing overt ventriculomegaly in adults. J Neurosurg 2000;92:933–40. [17] Eide PK, Sorteberg W. Association among intracranial compliance, intracranial pulse pressure amplitude and intracranial pressure in patients with intracranial bleeds. Neurol Res 2007;29:798–802. [18] Bering Jr EA. Circulation of the cerebrospinal fluid. Demonstration of the choroid plexuses as the generator of the force for flow of fluid and ventricular enlargement. J Neurosurg 1962;19:405–13. [19] Ambarki K, Israelsson H, Wahlin A, Birgander R, Eklund A, Malm J. Brain ventricular size in healthy elderly: comparison between Evans index and volume measurement. Neurosurgery 2010;67:94–9 [discussion 9].

Association between ventricular volume measures and pulsatile and static intracranial pressure scores in non-communicating hydrocephalus.

In non-communicating hydrocephalus (HC), enlarged cerebral ventricles are often thought to reflect increased intracranial pressure (ICP) or increased ...
824KB Sizes 0 Downloads 11 Views