Saadoun and Papadopoulos Critical Care (2016) 20:308 DOI 10.1186/s13054-016-1490-3 VIEWPOINT Open Access Spinal cord injury: is monitoring from the injury site the future? Samira Saadoun and Marios C. Papadopoulos* within 24 h compared with >24 h after cervical TSCI Abstract [4]. This study was underpowered, not randomized, and This paper challenges the current management of not blinded. In the UK [3] and internationally [2, 5] acute traumatic spinal cord injury based on our there is no consensus on the timing or even the role of experience with monitoring from the injury site in the surgery for TSCI. Below, we argue that early surgery is neurointensive care unit. We argue that the concept controversial because surgeons perform bony decom- of bony decompression is inadequate. The concept of pression, but fail to relieve the dural compression. optimum spinal cord perfusion pressure, which differs between patients, is introduced. Such variability Medical and nursing management suggests individualized patient treatment. Failing to There are no drugs that improve outcome after TSCI. The optimize spinal cord perfusion limits the entry of North American Spinal Cord Injury Studies suggested that systemically administered drugs into the injured cord. methylprednisolone given within 8 h after TSCI improves We conclude that monitoring from the injury site outcome [6, 7], but their findings have been criticized; helps optimize management and should be subjected methylprednisolone is no longer standard of care [3, 8, 9]. to a trial to determine whether it improves outcome. The optimum mean arterial pressure (MAP) after TSCI is Keywords: Blood pressure, CNS injury, Clinical trial, unknown. The American Association of Neurological Microdialysis, Monitoring, Neurocritical care, Spinal Surgeons recommends a MAP of 85–90 mmHg for 7 days cord injury, Surgery with little supporting evidence [10]. In the UK, blood pressure management is variable with 23 % of NICU spe- cialists aiming for MAP >60 mmHg, 54 % >80 mmHg, Background and 15 % within 20 % of what is considered normal for – Every year, 15 40 people per million suffer a traumatic age [3]. The effects on the injured cord of different anes- spinal cord injury (TSCI) [1]. Many TSCI patients are thetics, altering arterial pCO2, administering vasopressors initially admitted to a neurointensive care unit (NICU), or mannitol, and patient position in bed are also unclear. but their management is variable. This review introduces In the UK, 96 % of neuroanesthesiologists avoid N2O[3] novel concepts to aid the management of acute TSCI in which elevates intracranial pressure (ICP), though its NICU based on our findings that: 1) the dura causes effect on intraspinal pressure (ISP) is unknown. Most spinal cord compression at the injury site; 2) each pa- surgeons fix the spine and mobilize patients early [2], but tient has an optimum spinal cord perfusion pressure; in the UK Midlands Spinal Injuries Centre patients were and 3) patient position in bed influences cord perfusion. kept flat for 6 weeks after TSCI with an equally good out- comes [11]. Such variability in practice suggests that the Patient management optimum medical and nursing management are unknown. Surgical management Below, we show how neuromonitoring may be used to Some surgeons recommend early bony decompression, optimize management. but others do not [2, 3]. The Surgical Timing in Acute Spinal Cord Injury Study showed better outcome at How to optimize management 6 months in patients who had decompressive surgery History of monitoring from the injury site In 2009, whilst studying the role of the water channel * Correspondence: [email protected] Academic Neurosurgery Unit, St. George’s, University of London, Cranmer protein AQP4 in TSCI in mice, Dr Saadoun noticed that Terrace, Tooting, London SW17 0RE, UK the injured cord was compressed against the dura, based © 2016 The Author(s). Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Saadoun and Papadopoulos Critical Care (2016) 20:308 Page 2 of 9 on myelograms and ISP measurements [12]. She spine is fixed, it is difficult to see on MRI. For these rea- wondered whether the same occurs in TSCI patients. To sons, ISP has not attracted the same interest as ICP. investigate this, we set up a clinical trial termed Injured Spinal Cord Pressure Evaluation (ISCoPE), aiming to Technique of monitoring from the injury site monitor ISP from the injury site in humans. By 2014, we The probe is inserted intradurally during surgery at the had monitored the ISP in 18 patients with severe TSCI injury site to monitor ISP for a week (Fig. 1a). The probe and had shown that ISP is high and potentially detri- cannot be inserted in the NICU, thus precluding moni- mental [13]. A detailed morphological and spectral ana- toring non-surgical patients. ISP monitoring is tech- lysis of the ISP signals followed [14]. Our studies led to nically simple and safe based on data from 42 patients the idea of dural spinal cord compression that causes [19]. The probe is easily removed in the NICU. compartmentalization at the injury site [15–17]. We The ISP and ICP waveforms are similar with three thus evaluated the effect of duroplasty after TSCI [18]. characteristic peaks, identical Fourier transforms, and In 2016, we reviewed all ISCoPE patients and demon- identical shape change as ISP/ICP rise [14] (Fig. 1b). strated the safety and probe placement accuracy of the Thus, mathematically, the ISP signal can be subjected to technique [19]. Multi-modality monitoring from the the same analyses as the ICP signal [25]. Several parame- injury site was introduced in the same year [20], in- ters can be computed from these signals that provide cluding a novel analysis technique using Kohonen self- physiological information about the injured cord (Table 1). organizing maps [21]. The ISCoPE studies are ongoing. It is important to note that we monitor ISP from the sub- dural space, which is less invasive than intraparenchymal Brain versus spinal cord injury ISP. When the injured cord is swollen and compressed Neuromonitoring is the standard of care for acute severe against the dura, subdural ISP equals intraparenchymal traumatic brain injury (TBI). We use the term neuromo- ISP at the injury site [13, 17, 26]. After a mild TSCI, the nitoring to mean monitoring ICP, tissue oxygenation, spinal cord may not be compressed against the dura; in etc., in the NICU, rather than intraoperative neuro- this case, the relation between subdural versus intrapar- physiological monitoring. Though a recent study sug- enchymal ISP is unknown. To date, we have only moni- gested that ICP monitoring does not improve outcome tored ISP from patients with severe TSCI. [22], the findings are confounded by a delay in the start It is also possible to perform multi-modality monitor- of monitoring and no rehabilitation after discharge from ing by placing ISP and MD probes on the spinal cord the intensive care unit. Systematic reviews suggest that surface at the injury site [20]. In our recent study of 14 ICP monitoring in TBI reduces mortality [23] and in- TSCI patients, we found that surface MD monitoring is creases the chance of a favorable outcome [24]. There safe [20]. There was severe metabolic derangement at are advocates of multi-modality monitoring from the in- the injury site that correlated with the severity of injury. jury site including microdialysis (MD), oxygen, blood Below, we discuss how MD can be used to determine flow, and electrocorticography based on the idea that the optimum spinal cord perfusion pressure (SCPPopt) secondary damage is multifactorial and arises not only and optimum tissue glucose concentration as well as from ischemia, but also from metabolic disturbance and maximize the penetration of systemically administered spreading depolarizations. Probes are routinely intro- drugs into the injury site. duced into the injured brain, but no such monitoring exists for TSCI. Animal models Several reasons may account for the lack of neuromo- After TSCI in mice, the injured cord swells and is com- nitoring in TSCI. First, there is concern that inserting pressed against the dura thus generating high ISP [12]. probes exacerbates spinal cord damage. Second, spinal Mice that lack the water channel protein aquaporin-4 probes require surgery for insertion. Third, many TSCIs have reduced spinal cord edema after TSCI with im- are managed by orthopedic surgeons not accustomed to proved outcome. Reducing the elevated ISP in a rat neuromonitoring. Fourth, the clinical impact of elevated contusion TSCI model also improved outcome [27]. In a ICP is easy to detect in anesthetized brain-injured pig contusion TSCI model, increased thecal sac dimen- patients as a loss of pupillary light reflex, dilated pupil, sions were associated with reduced cord compression and cardiovascular instability, and, ultimately, brain death. In reduced injury site ISP [28]. Durotomy reduced intrapar- contrast, increased ISP is difficult to detect in anesthe- enchymal ISP after TSCI in ex-vivo pig spinal cords [29]. tized TSCI patients. This requires monitoring of motor In a rat contusion TSCI model, durotomy with duroplasty or sensory evoked responses, which is not routinely improved outcome more than durotomy without duro- done. Fifth, CT scans are routinely used in TBI to decide plasty [27]. After rat TSCI, durotomy without duroplasty management, e.g., decompressive craniectomy. In con- was associated with more macrophage accumulation, cys- trast, the injured cord is invisible on CT and, once the tic cavitation, and fibroblast proliferation than durotomy Saadoun and Papadopoulos Critical Care (2016) 20:308 Page 3 of 9 Fig.