Abstract
Purpose
Intrathecal baclofen (ITB) therapy is an accepted treatment modality for spasticity and dystonia. Several complications related to ITB have been described, including mechanical malfunctions, infections, cerebrospinal fluid fistula, and baclofen withdrawal or overdose. In this study, we present our institutional experience with ITB therapy, emphasizing complication avoidance and lessons learned.
Methods
The charts of 87 patients treated with ITB therapy were retrospectively reviewed. The primary surgical technique, complication type and timing, method of treatment, and outcome were analyzed.
Results
Thirteen out of 76 (17.1%) patients primarily treated at our department had 25 complications. The first complication occurred 17.5–30.9 months (mean 24.2 ± 6.7) after the pump implantation. Additional four patients with pumps placed elsewhere had six complications and were subsequently treated by our group. The main complications were: catheter fracture (11), subcutaneous fluid collection (5), lumbar wound/CSF infection (3), lumbar catheter or connector protrusion (3), pump malfunction (3), distal catheter migration outside the thecal sac (2), and baclofen withdrawal (1). Of the patients in the NYULMC group, six were treated by a single surgical procedure, six underwent multiple surgical procedures, and one was managed conservatively. In retrospect, changing the surgical technique, or adding an abdominal binder may have prevented 17 complications (54.8%). There were two deaths that were unrelated to the ITB therapy.
Conclusion
ITB therapy is associated with complications, many of which require additional surgery. Some of these complications are avoidable by adhering to a strict surgical technique and a proper criterion for patient selection.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Intrathecal baclofen (ITB) systems have been effectively used since the mid-1980s for treating patients with severe spasticity [29]. The therapeutic advantage of ITB systems has especially been seen in cases with medically refractory spasticity. Common indications for ITB therapy include cerebral palsy, traumatic brain injury, spinal cord injury, diffuse anoxic brain injury, hereditary diseases such as Rett syndrome, and other etiologies leading to severe and unmanageable spasticity [17, 19, 25]. Several studies have supported the use of ITB systems for the treatment of dystonia too [5, 27]. Although some patients may not regain complete functionality or remission of pain, ITB therapy has shown to positively affect the quality of life and ease patient daily treatment by their caregivers.
While ITB is an effective treatment option, there are still significant complications associated with the intricate surgical technique and prolonged duration of treatment. Complications may range from constitutional symptoms related to the Baclofen drug (such as diarrhea or vomiting), to equipment failure, and infections [5, 31].
In this study, we present our institutional experience with ITB therapy, emphasizing complication avoidance and lessons learned.
Methods
All cases undergoing ITB pump-related procedures at the Department of Neurosurgery at New York University Langone Medical Center (NYULMC) between July 1998 and December 2008 were included in this series. Data were retrospectively collected from the surgical and follow-up charts at the Pediatric Neurosurgical Division at NYULMC, as well as from the preoperative and follow-up charts at the Hospital of Joint Disease, NY, NY. IRB approval was given prior to data collection.
The current series includes two subgroups: patients that underwent primary implantation of an ITB pump at the NYULMC, and those who were implanted elsewhere and were treated for ITB-related complications at our center. As a routine, all patients who underwent a primary implantation of ITB pumps were referred by a neurologist specializing in spasticity, rehabilitation, or movement disorders. The patients underwent presurgical neurological and physical therapy evaluations, and were followed by the same neurologist postoperatively. Pre- and postoperative Ashworth spasticity scores were available for only a subset of patients, and as the focus of this paper was not the degree of spasticity improvement, we did not include the scoring in the current paper.
The goals for ITB therapy in the current series were increasing the functionality of patients who were cooperative and awake, as well as palliative treatment, aiding the primary caregivers in the routine treatment of patients who were not cooperative or in chronic vegetative state (such as following diffuse anoxic brain injury).
ITB pump implantation was routinely preceded by an ITB trial, and only patients responding favorably to the trial were offered a permanent pump.
Implantation procedure
The site of pump placement is decided after discussion with the patient and/or their caretaker about their daily activities, posture of the body, and previous procedures such as gastrostomy, VP shunt, etc. Following preoperative prophylactic antibiotics (typically Ancef or Clindamycin), patients undergo general endotracheal anesthesia and are placed in either a left or right lateral decubitus position depending upon the site of the pump placement. For placement of the intrathecal catheter, a Touhy needle is used to enter the lumbar thecal sac through a vertical incision at the level of the lower lumbar spine (L4-L5 or L3-L4). The needle is then angulated upward. As a routine, fluoroscopy is not used; however, for select patients, such as those with spinal fusion, or for patients with scoliosis, fluoroscopy is recommended, and rarely, a laminectomy may be necessary to access the spinal canal. Upon entering the lumbar cistern, the catheter along with guidewire is advanced to the planned spinal level (approximately lower thoracic spine in paraparetic cases, and upper thoracic spine in tetraparetic cases). The height of the intrathecal catheter is estimated by presurgical measurement of the distance to the desired level. The subcutaneous tissue is dissected until the fascia is reached and a purse-string suture is placed around the needle, which is then withdrawn. The guidewire is then withdrawn too and spontaneous cerebrospinal fluid (CSF) flow is noted. The purse-string suture is then tied around the intrathecal catheter, thus preventing a pericatheter CSF leak. The intrathecal catheter is cut so that only about 4–5 cm of stump remains, and is attached to a straight connector and to the primary tube (that later attaches to the pump). A silicone covering is placed over the connection site and secured with ties. The primary tube is tunneled from the abdominal wall wound (at the level of the pump site—either subcutane, or subfascia) to the spinal subcutaneous pocket, and pulled so that the tubing lies loose with a small loose loop in the pocket, and the connector lies in the subcutaneous tract slightly lateral to the spinal midline incision.
For pump placement, a 10-cm horizontal skin incision is made slightly under the level of the lower rib. The subcutaneous tissue is then divided to create a pocket. If a subfascial pocket is desired, then the rectus sheath is identified and divided horizontally to create a subfascial pocket [24]. It is important to verify hemostasis prior to placement of the pump. All pumps implanted were Medtronic Synchromed EL or Synchromed II pumps, including 10, 18, 20, and 40 cc volumes. The pump is soaked in bacitracin solution and filled with baclofen. It is then connected to the primary catheter, a tie is placed over the connection site to protect against detachment, and the pump is placed in the pocket. Careful attention is paid to avoid any kinking or malposition of the catheter. The pump is placed such that the tubing lies posterior to the pump to prevent inadvertent tubing injury when filling the pump. Fascia, subcutaneous tissue, and skin are then closed in layers. Lastly, the pump is programmed to reflect catheter lengths, pump volume, and required daily dose.
Postoperative course
Following surgery, antibiotics are continued for 24 h, and patients are referred to rehabilitation or discharged home within 1–2 days.
Data analysis
Data was tabulated using Excel software. Data is presented as mean ± SD.
Results
A total of 87 patients were included in the current series. There were 59 males and 27 females between the ages of 5 and 27 years (16 ± 10). Of these patients, 76 received their first ITB pump implant at NYULMC (NYULMC group), and 11 were referred to NYULMC for an ITB pump-related complication after being primarily treated elsewhere (non-NYULMC group). The indications for ITB pump are presented in Table 1. Seventy-seven (88.5%) patients were treated with the goal of functional improvement, while therapy for ten (11.5%) patients was directed as palliative care. Thirty-four (39%) patients underwent surgical procedures for non-complicated indications such as battery expiration/pump size upgrade (29), pump removal (3), and catheter revision for spinal surgery (2). No major intra-operative complications occurred.
Complications
There were a total of 25 complications noted in 13 (17%) patients primarily treated at NYULMC. The first complication in all patients occurred between 17.5 and 30.9 months (24.2 ± 6.7) after pump implantation. Four patients in the non-NYULMC group had a total of six complications, which were treated at our institution. Complications included catheter fracture (11), subcutaneous fluid collection (5), lumbar wound/CSF infection (3), lumbar catheter or connector protrusion (3), pump malfunction (3), distal catheter migration outside the thecal sac (2), catheter disconnection (1), intrathecal catheter flipping (1), baclofen withdrawal (1), and idiopathic system malfunction (1; Table 2).
Amongst the NYULMC group, one patient was treated conservatively with no surgical treatment, six patients were treated successfully with one surgical procedure, and the remaining six patients required multiple surgical procedures. Of the four patients in the non-NYULMC group, two patients were treated with one surgical procedure, one patient was treated with two surgical procedures, and one patient underwent one surgical revision for a complication at NYULMC and a separate revision for another complication at another institution.
Preventable complications
Of the 25 complications in the NYULMC group of patients, 17 complications in a total of nine patients were presumed as being preventable. None of the six complications noted in the non-NYULMC group were attributed to preventable factors. In these scenarios, stricter criteria for patient selection, emphasis on postoperative care, or a re-envisioned surgical technique are thought to be the key in the prevention of such complications in the future.
Two patients (nos. 4 and 13) suffered multiple catheter-related complications due to severe dystonia. Both patients experienced fluid collections resulting from catheter fractures as well as a secondary complication due to either migration of the catheter outside the thecal sac or disconnection of the catheter. These complications were attributed to the fact that both patients were noted to be severely writhing due to excessive truncal dystonic movements. To avoid catheter malfunction in severely dystonic patients, intraventricular baclofen should be considered.
Two patients experienced catheter-related complications that could have been prevented with use of alternate surgical technique. One patient (no. 8) was initially treated for a complication at another institution, but came to NYULMC for revision of a spinal fluid collection. Upon exploration of the hardware, it was noted that there was no purse-string suture at the catheter exit site. This enabled the catheter to coil and migrate inside the thecal sac and cause the fluid collection. In the second patient (no. 12), the suture at the catheter–pump connection site had eroded through the catheter. This complication could have been prevented with the use of a sutureless silicon cap at the catheter–pump connection.
Equipment protrusions were preventable too, and were seen in three patients (nos. 6, 7, and 9). Protruding equipment included the pump (one patient) and the spinal right angle fixator (two patients). All three patients were required to undergo surgical revisions and removal of equipment to treat their complications (removal of right angle fixator in the two cases, and removal of all hardware in the third). In retrospect, the cause of protrusions was attributed to either the right angle fixator or subcutaneous pump placement. Such complications can be avoided by avoiding use of the right angle fixator and implanting pumps subfascially especially in cachectic patients.
Of the preventable fluid collections seen in two patients—one was treatable by replacement of the abdominal binder (no. 1), while the other required several surgical revisions and eventually equipment removal due to hardware infection (no. 10). In the second case, recurring pseudomeningoceles were related to an increased CSF pressure and moderate hydrocephalus. We believe that early CSF diversion may have obviated the need for multiple wound revisions leading to further complications.
Discussion
Intrathecal baclofen is widely used for treating spasticity, increasing the functionality in these patients and their overall quality of life. While the efficacy of ITB therapy has been supported by various studies [2, 7, 18, 23, 26, 32], there is still a wide range of complications associated with the treatment. Approximately 20–30% of patients with baclofen pumps have complications [14, 22], these include baclofen-related complications (overdose and withdrawal), infections (surgical site infections, hardware infections, and meningitis), and hardware-related complications (malfunction, leaks, etc). The most frequent complications include infections, CSF leaks, and catheter malfunctions [7, 31, 32]. Rarely, baclofen withdrawal and overdose can also occur [9, 12, 19].
Upon revision of our patient series, we noted that some complications could have been prevented by meticulous care in patient selection, surgical technique, and postoperative follow-up. These complications can broadly be classified into: catheter-related (migration or disconnection) and equipment protrusion, CSF leaks and infection.
Catheter migration or disconnection
Catheter-related complications are the most common device-related complications [13, 33]. Many studies have stated that the majority of complications that required repeated surgeries involved the catheter, both at the thecal and the pump site [1, 3, 11, 16, 20, 31]. In the current series, several cases had catheter-related complications. As noted in the literature, these complications were recurrent [1, 3, 11, 16, 20, 31]. The most common presenting sign was spinal/abdominal fluid collections. These complications were seen to be a result of the patients’ severe dystonia and truncal rotation. We recommend that patients suffering from severe dystonia and contortion should be counseled of the higher possibility of catheter-related complications. Previous studies have shown that intraventricular baclofen can be effective in treating patients with secondary as well as heterodegenerative dystonia, the site of action being at the cortical level [5, 8]. Intraventricular baclofen should be offered to the patients who primarily present with severe dystonia. We also observed certain instances where the fluid collection was caused by the securing tie eroding through the catheter connection to the pump. Currently, there are sutureless catheter connectors available that can be used to prevent these complications. We also had cases where the lumbar fluid collection was due to lack of purse-string suture at the catheter exit site. Purse-string suture at the level of the fascia is important also to prevent extrathecal slippage and coiling of the catheter.
Equipment protrusion
Equipment protrusion is a rare complication [28, 30, 31]. We observed equipment protrusion including exposure of the pump or the right angle spinal fixator in three cases. All three patients were extremely thin. A subfascial implantation in very thin patients to allow greater tissue coverage of the equipment could have prevented this [24]. Additionally, placement of the right angle fixator above the spinal fascia does not seem to be important, and may be omitted. Currently, we place the right-angle adaptor only in patients with thick subcutaneous tissue, in which the surgical wound is large and additional fixation is needed.
Abdominal pocket fluid collection
Postoperative effusions are one of the less commonly seen complications of baclofen pump surgery [3, 14, 22, 26, 31]. It may be benign (such as a seroma), or due to under lying problems like CSF leak, catheter-related issues, or infection. If the patient does not present with any other symptom other than the fluid collection, it is best observed and abdominal binder applied. If the effusion persists, then further investigation such as fluid aspiration (under strict sterile conditions), and imaging (such as a CT scan following contrast injection through the pump—to exclude a leak) should be done. We had one patient who presented with fluid collection in the abdominal pocket. This was due to premature removal of the abdominal binder. We recommend wearing an abdominal binder for at least 6 weeks after implantation of the ITB pump system to prevent fluid collection.
CSF leaks
CSF leakage is one of the common complications encountered in ITB pump implantation [3, 13, 22, 26, 31]. CSF leaks may cause pseudomeningocele and impair wound healing, and may lead to an infection if the wound breaks down [10, 13, 22, 24]. We observed one patient with an external CSF leak. In this case, there was a large pseudomeningocele that developed 6 months after the ITB pump implantation. A local thecal catheter and wound revision was followed by external CSF leaks. This patient had moderate ventriculomegaly, and a high opening lumbar pressure. Eventually, this patient got infected, and the hardware was removed. Finally, a ventriculoperitoneal shunt was placed due to onset of symptoms of elevated intracranial pressure (ICP). “Occult hydrocephalus” was found in a subset of patients with spasticity, despite the absence of symptoms of elevated ICP [4, 6]. This may lead to CSF leaks and pseudomeningoceles. We recommend that CSF opening pressures should be routinely measured prior to implantation of the ITB pump (for example, during the baclofen test), and brain imaging should be performed to rule out ventriculomegaly. We also recommend that a ventriculoperitoneal shunt should be considered in patients with high CSF opening pressures and ventriculomegaly prior to ITB pump implantation, even in the absence of overt symptoms and signs of elevated ICP.
Infections
Reported rates of infection vary between 3.4% and 41.7%, and may include wound infections and meningitis [3, 10, 11, 13, 15, 21, 22, 31, 34]. Two cases in the current series had superficial wound infections. One of the cases was secondary to external CSF leak, and the other was due to equipment protrusion. Both infections were secondary to other complications. Thus, every effort should be done to avoid the primary complications, which in turn decreases the risk of infection.
Study limitations
The retrospective nature, lack of pre- and postoperative Ashworth spasticity scores for some of the patients, and lack of details regarding patients treated elsewhere (despite it being a small group) make this analysis incomplete. However, the study focuses on a single centers’ experience treating children with ITB, and thus, the lessons learned from the current series may shed light on the causes and prevention of some of these complications.
Conclusions
ITB therapy is an effective functional and palliative treatment for spasticity and dystonia caused by cerebral palsy, TBI, hereditary diseases, and other etiologies.
Although many complications are still associated with the treatment, careful analysis of such complications reveals preventative methods that can be taken to avoid some of the complications.
References
Albright AL, Barron WB, Fasick MP, Polinko P, Janosky J (1993) Continuous intrathecal baclofen infusion for spasticity of cerebral origin. JAMA 270:2475–2477
Albright AL, Gilmartin R, Swift D, Krach LE, Ivanhoe CB, McLaughlin JF (2003) Long-term intrathecal baclofen therapy for severe spasticity of cerebral origin. J Neurosurg 98:291–295
Albright AL, Awaad Y, Muhonen M, Boydston WR, Gilmartin R, Krach LE, Turner M, Zidek KA, Wright E, Swift D, Bloom K (2004) Performance and complications associated with the synchromed 10-ml infusion pump for intrathecal baclofen administration in children. J Neurosurg 101:64–68
Albright AL, Ferson S, Carlos S (2005) Occult hydrocephalus in children with cerebral palsy. Neurosurgery 56:93–96, discussion 96–97
Albright AL (2006) Intraventricular baclofen infusion for dystonia. Report of two cases. J Neurosurg 105:71–74
Albright AL, Turner M, Pattisapu JV (2006) Best-practice surgical techniques for intrathecal baclofen therapy. J Neurosurg 104:233–239
Albright AL (2007) Intrathecal baclofen for childhood hypertonia. Childs Nerv Syst 23:971–979
Albright AL, Ferson SS (2009) Intraventricular baclofen for dystonia: techniques and outcomes. Clinical article. J Neurosurg Pediatr 3:11–14
Alden TD, Lytle RA, Park TS, Noetzel MJ, Ojemann JG (2002) Intrathecal baclofen withdrawal: a case report and review of the literature. Childs Nerv Syst 18:522–525
Armstrong RW, Steinbok P, Cochrane DD, Kube SD, Fife SE, Farrell K (1997) Intrathecally administered baclofen for treatment of children with spasticity of cerebral origin. J Neurosurg 87:409–414
Awaad Y, Tayem H, Munoz S, Ham S, Michon AM, Awaad R (2003) Functional assessment following intrathecal baclofen therapy in children with spastic cerebral palsy. J Child Neurol 18:26–34
Bellinger A, Siriwetchadarak R, Rosenquist R, Greenlee JD (2009) Prevention of intrathecal baclofen withdrawal syndrome: successful use of a temporary intrathecal catheter. Reg Anesth Pain Med 34:600–602
Borowski A, Shah SA, Littleton AG, Dabney KW, Miller F (2008) Baclofen pump implantation and spinal fusion in children: techniques and complications. Spine (Phila Pa 1976) 33:1995–2000
Borowski A, Littleton AG, Borkhuu B, Presedo A, Shah S, Dabney KW, Lyons S, McMannus M, Miller F (2010) Complications of intrathecal baclofen pump therapy in pediatric patients. J Pediatr Orthop 30:76–81
Boviatsis EJ, Kouyialis AT, Boutsikakis I, Korfias S, Sakas DE (2004) Infected CNS infusion pumps. Is there a chance for treatment without removal? Acta Neurochir (Wien) 146:463–467
Campbell WM, Ferrel A, McLaughlin JF, Grant GA, Loeser JD, Graubert C, Bjornson K (2002) Long-term safety and efficacy of continuous intrathecal baclofen. Dev Med Child Neurol 44:660–665
Coffey JR, Cahill D, Steers W, Park TS, Ordia J, Meythaler J, Herman R, Shetter AG, Levy R, Gill B et al (1993) Intrathecal baclofen for intractable spasticity of spinal origin: results of a long-term multicenter study. J Neurosurg 78:226–232
de Lissovoy G, Matza LS, Green H, Werner M, Edgar T (2007) Cost-effectiveness of intrathecal baclofen therapy for the treatment of severe spasticity associated with cerebral palsy. J Child Neurol 22:49–59
Douglas AF, Weiner HL, Schwartz DR (2005) Prolonged intrathecal baclofen withdrawal syndrome. Case report and discussion of current therapeutic management. J Neurosurg 102:1133–1136
Gerszten PC, Albright AL, Barry MJ (1997) Effect on ambulation of continuous intrathecal baclofen infusion. Pediatr Neurosurg 27:40–44
Gerszten PC, Albright AL, Johnstone GF (1998) Intrathecal baclofen infusion and subsequent orthopedic surgery in patients with spastic cerebral palsy. J Neurosurg 88:1009–1013
Gooch JL, Oberg WA, Grams B, Ward LA, Walker ML (2003) Complications of intrathecal baclofen pumps in children. Pediatr Neurosurg 39:1–6
Hoving MA, van Raak EP, Spincemaille GH, Palmans LJ, Becher JG, Vles JS (2009) Efficacy of intrathecal baclofen therapy in children with intractable spastic cerebral palsy: a randomised controlled trial. Eur J Paediatr Neurol 13:240–246
Kopell BH, Sala D, Doyle WK, Feldman DS, Wisoff JH, Weiner HL (2001) Subfascial implantation of intrathecal baclofen pumps in children: technical note. Neurosurgery 49:753–756, discussion 756–757
Korenkov AI, Niendorf WR, Darwish N, Glaeser E, Gaab MR (2002) Continuous intrathecal infusion of baclofen in patients with spasticity caused by spinal cord injuries. Neurosurg Rev 25:228–230
Motta F, Buonaguro V, Stignani C (2007) The use of intrathecal baclofen pump implants in children and adolescents: safety and complications in 200 consecutive cases. J Neurosurg 107:32–35
Narayan RK, Loubser PG, Jankovic J, Donovan WH, Bontke CF (1991) Intrathecal baclofen for intractable axial dystonia. Neurology 41:1141–1142
Patterson V, Watt M, Byrnes D, Crowe D, Lee A (1994) Management of severe spasticity with intrathecal baclofen delivered by a manually operated pump. J Neurol Neurosurg Psychiatry 57:582–585
Penn RD, Kroin JS (1984) Intrathecal baclofen alleviates spinal cord spasticity. Lancet 1:1078
Stetkarova I, Yablon SA, Kofler M, Stokic DS (2010) Procedure- and device-related complications of intrathecal baclofen administration for management of adult muscle hypertonia: a review. Neurorehabil Neural Repair 24(7):609–619
Vender JR, Hester S, Waller JL, Rekito A, Lee MR (2006) Identification and management of intrathecal baclofen pump complications: a comparison of pediatric and adult patients. J Neurosurg 104:9–15
Ward A, Hayden S, Dexter M, Scheinberg A (2009) Continuous intrathecal baclofen for children with spasticity and/or dystonia: goal attainment and complications associated with treatment. J Paediatr Child Health 45:720–726
Williams BS, Christo PJ (2009) Obstructed catheter connection pin discovered during intrathecal baclofen pump exchange. Clin J Pain 25:256–259
Wunderlich CA, Krach LE (2006) Gram-negative meningitis and infections in individuals treated with intrathecal baclofen for spasticity: a retrospective study. Dev Med Child Neurol 48:450–455
Author information
Authors and Affiliations
Corresponding author
Additional information
Neil Haranhalli and Dhanya Anand have equally contributed to this paper.
Rights and permissions
About this article
Cite this article
Haranhalli, N., Anand, D., Wisoff, J.H. et al. Intrathecal baclofen therapy: complication avoidance and management. Childs Nerv Syst 27, 421–427 (2011). https://doi.org/10.1007/s00381-010-1277-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00381-010-1277-9