Supportive Care in Cancer

, 18:137

Intraspinal techniques for pain management in cancer patients: a systematic review

Authors

    • Palliative Care Consult Team, Odette Cancer CentreSunnybrook Health Sciences
  • Vincent Chan
    • Department of AnesthesiaUniversity Health Network
  • Virginia Jarvis
    • The Ottawa Hospital Regional Cancer Centre
  • Cindy Walker-Dilks
    • Program in Evidence-Based Care, Department of Clinical Epidemiology and BiostatisticsMcMaster University
Review Article

DOI: 10.1007/s00520-009-0784-2

Cite this article as:
Myers, J., Chan, V., Jarvis, V. et al. Support Care Cancer (2010) 18: 137. doi:10.1007/s00520-009-0784-2

Abstract

Purpose

This systematic review outlines current evidence regarding the effectiveness of intraspinal techniques for cancer pain and addresses practical implementation issues.

Methods

A search of electronic databases identified systematic reviews and randomized controlled trials (RCTs) evaluating the effectiveness of intraspinal techniques in the setting of cancer pain. An environmental scan was completed via the internet to identify practice guidelines and resource documents addressing organizational and implementation issues in the delivery of intraspinal analgesia. Elements reviewed included patient selection, contraindications, monitoring, aftercare, follow-up, hospital discharge equipment, health personnel, patient education, and safety.

Main results

Three systematic reviews, three consensus conferences, and 12 RCTs met the inclusion criteria for evidence of effectiveness. No single systematic review or consensus conference included all relevant RCTs or specifically addressed the use of intraspinal techniques for cancer pain. Six RCTs compared intraspinal techniques alone or combined with other interventions alone or in combination, four compared different intraspinal medications, and two compared different intraspinal techniques. In general, the evidence supported the use of intraspinal techniques for cancer pain management. The two main indications consistently identified were intractable pain not controlled by other conventional medical routes and/or side effects from conventional pain management strategies preventing dose escalation. Reports indicate intraspinal analgesia is equally or more effective than conventional medical management and often associated with fewer side effects. Thirteen resource documents addressed issues surrounding the delivery of intraspinal analgesia and program implementation.

Conclusions

Intraspinal techniques monitored by an interprofessional health care team should be included as part of a comprehensive cancer pain management program.

Keywords

Intraspinal analgesiaCancer pain managementNeoplasms

Introduction

Through advances in knowledge and use of disease-modifying therapies, patients with cancer are living longer. Cancer-related pain, however, remains poorly managed [1]. In an effort to standardize and improve cancer pain management, the World Health Organization (WHO) endorsed a three-step approach outlining the use of opioids and adjuvant medications [2]. Because many of the cancer-related pain syndromes are complex, a significant number of patients continue to experience refractory pain despite adherence to the WHO guidelines [3]. In response to this, a revised approach was released by the WHO in 1996 identifying the fourth step known as “invasive therapy” [4]. Intraspinal (epidural or intrathecal) analgesia is an example of such a therapy.

Intraspinal catheters have been in routine use for decades in various surgical and anesthesia settings. In an effort to more specifically target pain pathways via the spinal cord, clinicians in the mid 1970s successfully introduced intraspinal catheter insertion and subsequent analgesic infusion as an effective intervention in the setting of refractory cancer pain. The delivery of intraspinal pain control is currently dependent on local policies and procedures and the interest and competence of interventional pain physicians, palliative care physicians, nurses, and family physicians in this field of care.

This review is intended to examine the clinical evidence related to intraspinal analgesia, as well as outline the resources required to support patient care before, during, and after intraspinal analgesia administration in the setting of cancer-related pain.

Methods

Evidence was selected and reviewed by a three-member working group derived from the Cancer Care Ontario Palliative Care Clinical Program that included a palliative care physician, a palliative care nurse, and an interventional pain physician.

To inform the evidence base supporting the clinical effectiveness of intraspinal pain techniques, MEDLINE (1950 to May, week 2, 2008), EMBASE (1980 to week 20, 2008), CINAHL (1982 to week 3, 2008), and the Cochrane Library were searched using medical subject headings and textwords pertaining to intraspinal, epidural, or intrathecal injections, pain and cancer, and combined with relevant study design terms and publication types for randomized controlled trials and systematic reviews.

In addition, an environmental scan was undertaken to search for clinical practice guidelines and resource documents (nonindexed) using the Google™ search engine on the internet. This scan was targeted at a list of organizations known to develop evidence-based practice guidelines of good quality or of relevance to the Ontario context. An untargeted search to find resource documents of relevant oncology, pain management, and anesthesiology professional organizations was also undertaken. Finally, the reference lists of all retrieved documents were scanned for additional relevant articles and working group members were asked about local guidelines, standards, and procedure manuals of which they were aware.

Study selection criteria

The following selection criteria were used to evaluate the clinical effectiveness literature:
  1. 1.

    study design: clinical practice guidelines, systematic reviews of RCTs, and RCTs

     
  2. 2.

    patient population: cancer patients (any age, any diagnosis) with cancer-related pain

     
  3. 3.
    intervention and comparisons:
    • intraspinal techniques alone or in combination versus (vs.) other interventions (e.g., medical management) alone or in combination

    • different intraspinal techniques

    • external pumps vs. internal pumps

    • timing of intraspinal techniques

     
  4. 4.

    outcomes: pain measured using a validated scale

     

Studies or reviews on the use of intraspinal techniques for procedure-related pain or techniques other than epidural or intrathecal analgesia (e.g., intracerebroventricular) were not included. Comments, letters, editorials, case reports, news, and non-English language articles were excluded.

To inform recommendations on clinical indications and organizational and implementation issues, clinical practice guidelines and resources documents (policy statements, position papers, practice parameters, care pathways, and manuals) were considered if they commented on at least one of: patient selection, contraindications, monitoring, aftercare, follow-up, hospital discharge, equipment, practice team, professional competencies, patient education, and patient safety.

Results

Systematic reviews, consensus conferences, and randomized controlled trials

Three systematic reviews [57], three consensus conferences [810], and 12 RCTs [1122] met the selection criteria of studies examining clinical effectiveness. Although the systematic reviews and consensus conferences included some relevant RCTs, no single systematic review or consensus conference included all relevant RCTs or sufficiently discussed the effectiveness of intraspinal analgesia for cancer-related pain to support recommendations. Further, while all included cancer patients, only two systematic reviews [6, 7] and one [9] consensus conference commented specifically on or presented separate data specifically relevant to this population. The remainder combined studies across all patient populations and did not provide specific cancer pain information.

Of the 12 RCTs, six compared intraspinal techniques alone or in combination with other interventions alone or in combination, four compared different intraspinal medications, and two compared different intraspinal techniques. Indicators of quality extracted from each study were publication status (full publication or meeting abstract), allocation concealment, blinding, statement of statistical power or sample size calculation, intention-to-treat analysis, and statement of sponsorship or funding.

Characteristics of the included systematic reviews, consensus conferences, and RCTs, including quality features, are in Tables 1, 2, and 3.
Table 1

Characteristics of systematic reviews and consensus conferences

Review

Question

Data sources

Study selection

Outcomes

# relevant RCTs

Raffaeli [6]

What is the recent evidence (1990 to 2005) on the use of intrathecal opioids for cancer and noncancer pain?

MEDLINE, reviewing journals, and searching pertinent internet sites

Studies on intrathecal administration of drugs for patients with chronic (cancer and noncancer) pain

Pain intensity, opioid dosage reduction, side effects, and device complications

3

Systematic review

Carr [7]

Among several questions, are different analgesic drug formulations and routes of administration associated with different patient preferences or different efficacy rates?

Update of the 2001 AHRQ report on management of cancer pain and 2002 conference on symptom management in cancer. Searched MEDLINE, Cochrane, CancerLit, and references of reviews

RCTs evaluating the efficacy of treatments for cancer-related pain

Pain

3

Systematic review

Stearns [9]

What are the benefits and practical management considerations of intraspinal analgesia for cancer pain?

Multidisciplinary expert panel consensus based on experience from participation in Polyanalgesic Consensus Conference 2003, Hassenbusch [8]

 

Role of intrathecal therapy, patient selection, delivery systems, contraindications and risks, best practice algorithm

1

Consensus conference

AHRQ Agency for Healthcare Research and Quality, RCT randomized controlled trial

Table 2

Study quality of RCTs meeting inclusion criteria

Study

Publication status

Allocation concealment

Blinding

Statistical power

Intention-to-treat analysis

Sponsorship funding

Comments and limitations

Vainio [11]

Full publication

NR

No

NR

Unclear

Finnish Cancer Society and Durascan Medical Products (implantable ports)

 

Smith [12]

Full publication

Yes

No

NR

Yes

Medtronic

<80% follow-up

Kalso [13]

Full publication

NR

Double

NR

Unclear

Academy of Finland; Paulo Foundation; Pharmacia (pump)

 

Staats [14]

Full publication

Yes

Double

96% to detect >30% change in VAS between groups (70 ziconotide and 35 placebo)

Yes

Neurex/Elan Pharmaceuticals and Medtronic

 

Eisenach [15]

Full publication

NR

Double

NR

NR

Fujisawa USA

<80% follow-up

Pasqualucci [16]

Full publication

NR

Double

NR

Unclear

NR

 

Dahm [17]

Full publication

NR

Double

NR

Unclear

Faculty of Medicine of Goteborg University and Inga-Britt and Arne Lundberg Research Foundation

<80% follow-up

van Dongen [18]

Full publication

NR

Double

NR

No

NR

Five patients were not randomized and received open-label treatment. Five patients in the morphine-alone group crossed over to combo Rx and were analyzed in the combo group.

Lauretti [19]

Full publication

Unclear

Double

Sample size calculation done—details not given

NR

Fundação de Amparo a Pesquisas de São Paulo

 

Yang [20]

Full publication

NR

Double

NR

Unclear

National Science Council, Republic of China

Pain intensity compared with pretrial, not between study groups

Gourlay [21]

Full publication

NR

No

NR

Unclear

Anti Cancer Foundation of the Universities of South Australia; Shiley Infusaid; Pharmacia (Port-a-Cath)

 

Georgiou [22]

Full publication

NR

No

NR

Unclear

NR

 

NR not reported, VAS visual analog scale

Table 3

Characteristics and results of included RCTs

Study

Patient characteristics, patient follow-up

Length of follow-up

Intervention

Outcomes

Results

Intraspinal techniques alone or in combination vs. other interventions alone or in combination

Vainio [11]

30 patients with severe cancer pain (mean age 52 years, range 23 to 86); follow-up, 30/30

To end of treatment (usually death of patient)

Epidural opiate using a conventional tunneled catheter (n = 10), implanted catheter, and injection port (n = 10), or systemic per oral opiate (n = 10)

Pain intensity (10-cm VAS), performance status (100-point [100 = normal] Karnofsky scale), side effects, and complications

Groups did not differ for VAS scores, which were <5 in all three groups. The oral group had more side effects than the epidural groups. Karnofsky scores were nonstatistically significantly better in the epidural groups than the oral group. Twelve incidences of complications occurred in the epidural groups compared with no complications in the oral group.

Smith [12] (intention-to-treat)

202 patients ≥18 years (mean age 57 years, 55% men) with advanced cancer and mean 10-point VAS (10 = worst) score ≥5; follow-up, 143/202 (71%)

6 months

Intrathecal morphine delivered through an implantable drug delivery system plus comprehensive medical management (n = 101) vs. comprehensive medical management alone (n = 99)

Clinical success, ≥20% reduction in VAS or equal VAS with ≥20% reduction in drug toxicity at 4 weeks

VAS pain score reduction 52% with implantable system vs.39% comprehensive medical management (p = 0.055). Clinical success 85% with implantable system vs.71% with comprehensive medical management (p = 0.05). Thirty patients in whom comprehensive medical management failed crossed over to implantable system and had significantly reduced VAS (27% reduction) and drug toxicity (51% reduction) scores within 4 weeks [38]. The as-treated analysis showed greater clinical success with implantable system than nonimplantable system (89% vs.71%, p = 0.02) [39].

Smith [38] (crossovers from comprehensive medical management)

Smith [39] (as-treated)

Kalso [13]

10 hospitalized patients (age range 22 to 75 years, 60% women) with severe cancer-related pain requiring opioids; follow-up, 9/10

Two 48-h periods

Crossover design: epidural or subcutaneous continuous infusion of morphine

A 100-mm VAS score (100 = worst) and adverse effects

Pain at rest was less during subcutaneous morphine than oral morphine (pretrial) and pain, while moving was less during subcutaneous and during epidural morphine than pretrial oral morphine; subcutaneous and epidural did not differ for either measurement. Pretrial oral morphine had more adverse effects than subcutaneous (P < 0.05). Epidural did not differ from either oral or subcutaneous for adverse effects.

Comparisons of different intraspinal administration techniques

Staats [14]

111 patients, 24 to 85 years (mean age 56 years, 50% men) with cancer (n = 95) or AIDS (n = 13) and mean 100-mm VAS (100 = worst) score ≥50; follow-up, 108/111 (97%)

Initial titration period 5 to 6 days plus a 5-d maintenance phase

Ziconotide (n = 71) or placebo (n = 40) by already implanted pump or by intrathecal catheter with external infusion. At end of titration, responders received 5 days of maintenance and nonresponders crossed over to alternate treatment.

Mean percentage change in VAS to end of titration phase

53% improvement with ziconotide vs.18% with placebo (p < 0.001)

Eisenach [15]

85 patients (age range 31 to 83 years, mean 56 years, 60% men) with severe cancer-related pain despite systemic or epidural morphine or morphine equivalents; follow-up, 56/85 (66%)

15 days

Continuous epidural infusion of clonidine, 30 μg/h (n = 38) or placebo (n = 47) by patient-controlled analgesia pump for 14 days; all patients had rescue epidural morphine.

Treatment success: any reduction in morphine use or VAS, or either variable decreasing with the other remaining constant

Treatment success was achieved by more clonidine patients than placebo patients (45% vs.21%, p = 0.016). Greatest success was seen in neuropathic pain control (50% vs.11%). Groups did not differ for adverse events (33% vs.33%).

Pasqualucci [16]

12 patients (age range 45 to 75 years, 75% men) with severe, continuous cancer-related pain (VAS 5/10) not responding to common analgesic drugs; follow-up, 12/12

18 h

Single dose via epidural of morphine, 3 mg (n = 6) or buprenorphine, 0.3 mg (n = 6)

Pain (10-cm VAS) and ventilatory function (breathing control and gas exchange)

Pain control improved in both groups through the first 6 h. Buprenorphine had significantly better pain control compared with baseline than morphine compared with baseline at 18 h.

Dahm [17]

21 patients 26 to 76 years (median age 63 years) with refractory pain conditions (15 cancer); follow-up:12/21 (9 cancer)

Two 7-d periods

Crossover design: continuous intrathecal infusion of ropivacaine or bupivacaine

Need for local anesthetics, mean 10-point VAS score (10 = worst), side effects

Need for local anesthetics higher with ropivacaine (62 vs.58 mg/day, P < 0.02), VAS scores no difference (P > 0.3), no difference in side effects

van Dongen [18]

20 patients (age range 35 to 82 years) with refractory cancer pain; follow-up, 20/20

Patients were followed until death (mean follow-up 58 to 85 days)

Intrathecal administration of morphine + bupivacaine (n = 11) or morphine alone (n = 9)

Progression of morphine dose and side effects

Adequate pain relief was achieved in both groups. The rate of dose progression of morphine was less in the morphine + bupivacaine group than the morphine alone group (0.0003 vs.0.005 mg/h, p = 0.0001).

Lauretti [19]

48 patients (mean age range 50 to 56 years, 63% men) with refractory cancer pain; follow-up, 48/48

25 daya

Epidural administration of morphine + ketamine, neostigmine, midazolam, or placebo (12 patients per group) whenever VAS score was ≥4/10

A 10-cm VAS (10 = worst), time to complaint of pain (VAS ≥ 4/10), and adverse effects

VAS scores 60 min after drug administration were lower with ketamine than midazolam (P = 0.018). Patients in the ketamine and neostigmine groups had a longer time to pain complaint than placebo patients. Groups did not differ for adverse events.

Yang [20]

20 hospitalized patients 22 to 69 years (50% men) with cancer using opioid analgesics for pain control; follow-up, 20/20

Two 48-h periods

Crossover design: intrathecal morphine + ketamine or morphine alone twice daily

Intrathecal morphine and rescue morphine requirements; pain intensity on 10-point VAS score (10 = worst); pain frequency on 4-point verbal ordinal scale (3 = constant)

The required dose of intrathecal morphine to be effective was less in the morphine + ketamine group (0.17 vs.0.38 mg, P < 0.05); pain intensity and pain frequency were both decreased after either intervention compared with pretrial.

Comparisons of different intraspinal medications

Gourlay [21]

29 Patients were followed until death (mean 140 to 169 days); follow-up 28/29

Epidural morphine administered as continuous infusion via pump (n = 15) or intermittent bolus via catheter (n = 14)

Epidural morphine administered as continuous infusion via pump (n = 15) or intermittent bolus via catheter (n = 14)])

VAS and problems encountered with the devices (0 to 4 [very troublesome])

VAS scores were low (good) in both infusion and bolus groups and did not differ, 1.48 vs. 1.23) and very few problems occurred with the devices (0.52 vs. 0.49).

Georgiou (2000) [22]

29 patients (mean age 68 y, 97% men) with terminal head or neck cancer with pain not controlled by oral morphine; follow-up, 28/29

Patients were followed until death; pain was measured up to 10 day.

Morphine administered epidurally in the thoracic (n = 16) or cervical region (n = 13).

100-mm VAS score (100 = worst), morphine use, and complications.

Groups did not differ in pain scores. The bolus dose and daily dose of morphine were lower in the cervical group (P < 0.05) and the duration of analgesia after each bolus dose was longer in the cervical group (P < 0.05). Nausea/vomiting and pruritus were mild or moderate and more common in thoracic patients, as was constipation.

h hour, mo month, VAS visual analog scale, wk week, y year

Clinical practice guidelines and resource documents

The environmental scan yielded 13 documents that were summarized and reviewed for quality and their utility in informing recommendations on the delivery of intraspinal pain management: eight documents were practice guidelines [2330], four were local internal use clinical care algorithms or policy and procedure documents [3134], and one was a practice standard [35]. The eight practice guidelines were evaluated using the Appraisal of Guidelines for Research and Evaluation instrument [36]. The guidelines varied in quality, but most could be recommended for use. The coverage of these resources for organizational guidance concerning intraspinal pain management is shown in Table 4.
Table 4

Organizational guidance: summary of evidence

Section

Practice Guidelines

Resource Papers: Policy and Standards

ONS 2004 [23]

BPS 2006 [24]

CCNS 2005 [25]

SIGN 2000 [26]

Sing 2003 [27]

APS 2005 [28]

NCCN 2007 [29]

ASA 1996 [30]

St Wilf 2001 [34]

Cal 2004 [31]

Corn 2001 [32]

Ott 2007 [33]

INS 2006 [35]

Indications for use

Patient selection

  

Contraindications

      

  

Practice setting

Monitoring, aftercare, follow-up

  

 

 

Hospital discharge

 

     

   

Equipment

      

 

 

Practice team

Practice team

     

  

 

Professional competencies

 

    

  

 

Patient education

  

 

  

Patient safety

      

  

Clinical effectiveness

Two systematic reviews met inclusion criteria [6, 7]. Raffaeli [6] reviewed the literature on intraspinal medication for chronic pain between 1990 and 2005. Of the 34 studies identified, 19 were on cancer pain and two of these were RCTs [12, 18]. Overall, the review demonstrated that intraspinal administration of opioids was effective in attenuating pain in patients with cancer pain and noncancer pain. A 2002 Agency for Healthcare Research and Quality evidence report [7] focused on the prevalence, assessment, and treatment of cancer-related pain, depression, and fatigue. RCT evidence was used to assess the efficacy of interventions. Intraspinal analgesia as an intervention was addressed in a summary of RCTs evaluating adjuvant analgesics. Three of the six RCTs were relevant [1719]. The review concluded insufficient RCT evidence was available to determine the relative efficacy of the spinal versus systemic route of drug administration.

Informed by a polyanalgesic consensus conference on the management of pain by intraspinal drug delivery across patient types [10], a similar multidisciplinary meeting was held to arrive at consensus-based recommendations for the use of intrathecal drug delivery for intractable cancer pain [9]. Based on consensus and citing the Smith 2002 RCT [12], the panel supported the use of intrathecal drug delivery in the management of cancer pain but noted that few RCTs have been conducted on the optimal medication for cancer pain, prompting reluctance among the medical community to accept intraspinal analgesia for widespread clinical use. The algorithm created by the polyanalgesic consensus conference was updated again in 2007, adding ziconotide to the list of first-line intrathecal medications [37].

The quality, characteristics, and results of the 12 eligible RCTs are in Tables 2 and 3. While most of the studies were double blinded, the reporting of the quality features was typically incomplete. Further, sample sizes varied widely, but most studies were small and likely underpowered.

Intraspinal techniques alone or in combination vs. other interventions alone or in combination

Three RCTs compared an intraspinal technique alone or in combination with another pain intervention alone or in combination. In a three-arm trial, Vainio [11] (n = 30) compared morphine given orally by conventional tunneled catheter or by implanted catheter. All three groups achieved marked pain relief and did not differ statistically from one another, but a higher rate of side effects was observed in the oral morphine group (16 occurrences compared with seven occurrences in the conventional epidural group and three occurrences in the port group). Technical complications were more common in the epidural groups and included dislocation in the epidural space (three times in the conventional catheter group), disconnection of the port (three times in the port group), and local infection of the skin (once in the conventional catheter group and twice in the port group).

Smith [12] (n = 202) compared intrathecal delivery of morphine through a programmable infusion system plus comprehensive medical management with medical management alone in 202 patients with refractory cancer pain. Clinical success was defined as ≥20% reduction in visual analog pain scale and ≥20% reduction in drug toxicity at 4 weeks. The difference between the two groups met borderline statistical significance (P = 0.05). The reduction in pain scores was greater in the intrathecal group, but the difference was just short of statistical significance (P = 0.055). Significant reductions in fatigue and depressed level of consciousness occurred in the intrathecal delivery group (P < 0.05). In two separately published reports, preplanned analyses of crossovers [38] and as-treated patients [39] showed statistically significantly greater clinical success with combined therapy.

Kalso [13] (n = 10) compared epidural with subcutaneous continuous infusion of morphine in a crossover trial. Both groups improved in pain control compared to that of oral morphine at baseline. The groups did not differ for adverse effects.

Comparisons of different intraspinal medications

Seven RCTs compared different types of drugs or compared drug with placebo in patients receiving an intraspinal drug delivery system. Staats [14] (n = 111) compared ziconotide with placebo by implanted pump or intrathecal catheter with external infusion. Patients who received ziconotide showed an improvement of 53% at the initial titration phase (5 to 6 days) compared with 18% improvement with placebo. Adverse effects such as somnolence, confusion, urinary incontinence, and fever were more common with ziconotide than placebo but were easily recognized and decreased when the dose of ziconotide was decreased. A long-term follow-up of open-label ziconotide [40] showed significant improvement in pain intensity scores from baseline at all time points and past 12 months.

Eisenach [15] (n = 85) compared a continuous epidural infusion of clonidine with placebo, both administered by patient-controlled pump. Significantly greater treatment success (defined as a decrease in morphine use or decrease in pain) was seen with clonidine than placebo (45% vs. 21%). Adverse events occurred in 33% of patients in both groups. Groups did not differ for sedation or decreased respiratory function, but clonidine decreased blood pressure by about 10 mmHg.

Pasqualucci [16] (n = 12) compared epidural administration of morphine with buprenorphine and showed significant reduction in pain in both groups at 6 h. At 18 h, better pain control (compared with baseline) was seen with buprenorphine in three pain indices compared with morphine in one pain index. Adverse reactions occurred in four of six buprenorphine patients and in two of six morphine patients.

In a crossover RCT comparing intrathecal bupivacaine with ropivacaine (n = 21), Dahm [17] showed no significant difference in pain scores but higher daily doses of local anesthetic to achieve a similar degree of pain relief were used with ropivacaine. Groups did not differ for side effects or complications.

Van Dongen [18] (n = 20) compared intrathecal delivery of morphine plus bupivacaine with morphine alone. Both groups had adequate pain relief, but the rate of morphine alone dose progression from day 10 to day 30 was greater in the morphine group than in the combined group (P < 0.001). No severe bupivacaine-related neurological deficits were observed.

Lauretti [19] (n = 48) evaluated patients allocated to epidural administration of morphine plus ketamine, neostigmine, midazolam, or placebo. Patients receiving ketamine had lower pain scores than those receiving midazolam at 60 min after administration. Pain relief lasted longer in the ketamine and neostigmine groups than the placebo group. Morphine consumption up to day 25 was lower in the ketamine group than the placebo group (P = 0.003). There were no differences among the groups for adverse effects, the most common being somnolence, constipation, diminished appetite, and skin redness around the epidural catheter and one patient experienced hallucinations.

In a crossover trial, Yang [20] (n = 20) compared intrathecal morphine plus ketamine with intrathecal morphine alone. Pain intensity and pain frequency were both decreased from baseline. The addition of ketamine resulted in a lower dose of morphine being required. Side effects did not differ between groups and were not serious. No cases of respiratory depression occurred.

Comparisons of different intraspinal administration techniques

Two small RCTs (each n = 29) compared different intraspinal techniques. Gourlay [21] evaluated epidural morphine administered as a continuous infusion by Infusaid pump or intermittent bolus by Port-a-Cath. Pain scores were low in both groups and did not significantly differ. Mean scores for problems associated with the devices were also low in both groups.

Giorgiou [22] evaluated epidural administration of morphine in the thoracic compared with the cervical region and showed no difference in pain scores. The cervical route required smaller bolus doses and gained longer lasting analgesia than the thoracic route. The thoracic route led to a greater rate of adverse effects (constipation, nausea and vomiting, and pruritus).

Indications for use

Among the documents that provided guidance on patient indications for the use of intraspinal analgesia [9, 2332], two indications were supported by all groups: intractable pain that could not be controlled by other conventional medical routes and side effects from conventional medical routes that prevented dose escalation. Table 5 outlines all the indications that were identified and the documents that supported those indications.
Table 5

Indications and contraindications for use of intraspinal analgesia for cancer pain

Category

ON[23]

BPS [24]

CCNS [25]

SIGN [26]

Sing [27]

APS [28]

NCCN [29]

ASA [30]

Cal [31]

Corn [32]

Stearns [9]

Indications for use of intraspinal analgesia

Intractable pain control

Dose-limiting side effects

Appropriate equipment and expertise available

  

 

   

  

Expectation that intraspinal analgesia would improve patient’s quality of life

   

   

  

Life expectancy of weeks or months

        

  

Informed consent

 

 

      

Contraindications

Bleeding diathesis

      

Active or local infection

       

Psychosocial distress

        

 

Unstable angina, severe left ventricular dysfunction, critical aortic stenosis, prohibitive spinal abnormality, and previous spinal fusion

        

  

Unfit for surgery or anesthesia

 

         

Disseminated disease contraindicated for regional therapy

         

Unstable vital signs, hematologic abnormalities, wound infections, emaciation, and spinal lesions

        

  

Requirement for large volume infusions

          

Presence of another implanted device

          

Five documents [9, 23, 24, 31, 32] addressed contraindications for the administration of intraspinal analgesia. Anticoagulation and active systemic or local infection were the most commonly reported contraindications. Upon reviewing, the working group felt that anticoagulation should be considered a contraindication to intraspinal analgesia at the time of catheter insertion, but not if the patient was anticoagulated after intraspinal analgesia was administered.

With respect to infectious considerations, septicemia is an absolute contraindication because of the high risk of catheter infection after implantation. This can lead to epidural abscess formation and the possibility of spinal cord compression resulting from a space-occupying abscess. This can also result in a CNS infection (e.g., meningitis and encephalitis). Sepsis or any systemic infection controlled after institution of antibiotic treatment is not a contraindication. Any local superficial infection after successful treatment is not a contraindication. An epidural/spinal catheter or an internalized pump may be implanted at a site or segmental level that is distant from the infected area. Patients with immunosuppression or insulin-dependent diabetes have a higher risk of infection.

Bleeding diathesis or any hemorrhagic conditions that can increase the risk of an epidural or spinal hematoma formation and long-term neurologic deficit is a contraindication. Patients on therapeutic or prophylactic anticoagulation treatment can receive intraspinal catheter treatment provided that anticoagulation is reversed prior to procedure. Similarly, patients with significant thrombocytopenia can receive intraspinal catheter treatment after adequate platelet transfusion. Patients with other coagulation disorders (e.g., von Willebrand disease, hemophilia) should be treated prior to procedure. Patients with known epidural metastases have a higher risk of spinal bleeding and hematoma if the catheter passes through the tumor mass. It is advisable to insert the catheter cephalad to the level of known or suspected spinal metastases.

Certain neurologic conditions were also considered contraindications for intraspinal pain control. These included unmotivated, noncompliant patients; preexisting uncontrolled or unstable CNS disorders (e.g., raised intracranial pressure); spinal canal pathology (e.g., spinal stenosis) that may predispose the patient to development of neurologic symptoms as a result of long-term spinal infusion and increased spinal canal pressure; and spine pathology (e.g., previous fusion or laminectomy) that may impair successful catheter placement. For the setting of active spinal cord compression, the working group felt careful consideration is warranted; however, this condition should not be considered a contraindication. Based on case reports and anecdotal experience, caution must be exercised with these patients as an evolving neurologic deficit may in fact represent either medication toxicity or a catheter placement related intraspinal bleed.

Seven documents [2325, 28, 3032] recommended a trial of intraspinal analgesia using a temporary epidural or spinal catheter to determine efficacy and appropriate dose range. One document [31] cautioned that limiting intraspinal analgesia to patients whose pain has failed to be controlled by other methods causes unnecessary waiting resulting in unacceptable suffering that leads to an even more complex pain management problem. Earlier consideration of intraspinal analgesia may lead to more rapid pain control and less patient suffering. The Polyanalgesic Consensus Conference recommended a screening trial before proceeding with pump implantation to evaluate patient response and potential side effects but indicated that a trial was not necessary for patients receiving external intraspinal systems [9].

Implementation issues for safe delivery of intraspinal analgesia

Several issues related to the safe delivery of intraspinal analgesia as it relates to equipment, aftercare, monitoring, hospital discharge, follow-up, practice team, professional education and competency, patient and family education, and patient safety were identified (Table 4).

Equipment

Specific equipment requirements for delivering intraspinal analgesia were included in six documents [9, 23, 24, 31, 33, 35]. None of the documents contained evidence comparing different types of delivery systems, but various options and the technical considerations in their use were presented. It is recommended that the choice of delivery system (e.g., epidural vs. intrathcal or implanted vs. external) be based on the HCP expertise at each institution factoring in any relevant practical considerations. Guidance covers general equipment requirements, as well as details concerning different kinds of intraspinal delivery systems: external, partially external, and fully implanted. Two issues concerning equipment that were considered particularly important in the administration of intraspinal analgesia were the use of preservative-free medication and the appropriate cleansing of the catheter site. The working group agrees with the INS [35] recommendations that straight alcohol or acetone should never be used for site preparation or cleansing but supports the use of staining disinfectants that contain alcohol because several studies comparing proviodine–iodine, chlorhexidine, and chlorhexidine with alcohol have found reduced epidural infection rates using chlorhexidine with alcohol as the skin disinfectant [4145].

Aftercare

Four documents provided advice concerning care to be given immediately after intraspinal analgesia insertion [23, 24, 31, 33]. Aftercare issues included the application of generic postoperative principles delivered by nurses trained in intraspinal drug delivery techniques, existence of appropriate protocols, availability of medical support and equipment, and protection of patients from any infectious environment (such as not being cared for on a ward where there are methicillin-resistant staphylococcus aureus patients). Immediate concerns after the procedure also included monitoring patients’ vital signs and pain levels; assessing the insertion site for such complications as leakage, bleeding, and infection; checking and changing the dressing; and assessing the proper functioning of the equipment.

Monitoring

Nine documents [9, 2325, 28, 3133, 35] provided guidance regarding monitoring of patients receiving intraspinal pain management. The careful assessment of pain control and monitoring for complications and side effects were emphasized. Many monitoring issues surrounding intraspinal patients pertain to the management of cancer-related pain in general. Tools and instruments for monitoring pain and detecting complications and adverse effects were found in three documents [25, 31, 33].

The inspection of intraspinal drug delivery devices was also emphasized as an important aspect of monitoring. Seven documents [9, 23, 24, 28, 31, 32, 35] warned of problems that could be encountered, including catheter malposition or migration, catheter disconnection, catheter occlusion (dislodgement, kinking, and erosion), and device failure or malfunction.

Hospital discharge

Four documents [24, 3032] provided advice regarding the hospital discharge of intraspinal patients. These documents stressed the importance of having a discharge plan in place before the initiation of the intraspinal analgesia procedure. The plans should ensure that properly trained professionals and an adequately equipped care setting are ready to receive the patient. In addition, the plans should include details about ordering equipment, drugs, and refills; the contact details of health professionals who will be involved in the patient’s care; and provision for the education of the patients, families, and home-care staff.

Follow-up

Six documents [23, 24, 3133, 35] provided advice for the follow-up of patients with intraspinal pain medication in the community. The interprofessional nature of follow-up care was emphasized, as was the importance of establishing protocols, so that all healthcare personnel are aware of the procedures to be followed in the intraspinal care of a patient, particularly in dealing with complications. Important elements for patient follow-up included ensuring the availability of pain management physicians outside of working hours to resolve problems that cannot be managed by nursing staff or local on-call medical staff. Such complications included increased sedation score, respiratory depression, signs of infection, persistent pain, and equipment malfunction. Provision should also be in place should patients require referral to hospital or hospice if their needs can no longer be met in the community. Similarly, if a patient moves away from the center where intraspinal analgesia was initiated, a plan should be in place to allow for the smooth transfer of care.

Practice team: interprofessional roles

The roles of healthcare professionals involved in cancer pain management emphasized the interprofessional nature of intraspinal care, the need for accurate communication among all persons involved, and the importance of working as a team that includes all healthcare professionals and patients and their families [23, 24, 30, 31, 34, 35]. Responsibilities should be in keeping with quality standards of relevant professional college or association certification. Participants in intraspinal care may include the referring healthcare professional (palliative care physician, advanced practice/specialist nurse, nurse practitioner, oncologist, or primary physician), interventional pain physician, registered nurse (palliative care unit, community, and inpatient), pharmacist (inpatient and community), and patient and family members.

Professional education and competencies

Six documents [23, 24, 26, 31, 34, 35] provided advice with respect to the necessary training and competencies for healthcare professionals involved in intraspinal care. Specific education training is required for healthcare professionals caring for patients receiving intraspinal analgesia. For example, the BPS [24] recommended a mentoring system among healthcare professionals and the establishment of a network of trained clinicians in order to provide coverage. Practitioners who do the implanting must have appropriate training and a caseload sufficient to maintain expertise. At this time, the ideal caseload is not known.

Nursing care of intraspinal patients is an advanced nursing competency requiring certification. Educational activities for nurses should cover the operation of intraspinal drug delivery systems, insertion and access procedures, care and maintenance (assessment of site, dressing change, and flushing and aspiration procedures), potential complications and interventions, and patient and family education. Routine continuing education opportunities to maintain knowledge and clinical competency are recommended [23, 31, 34, 35].

Patient and family education

Eight documents [2325, 28, 3033] included advice on patient and family education. The use of intraspinal analgesia in the community setting requires that patients and family members are frequently educated about and highly motivated and competent in the use of the drug delivery system. Documentation of what has been taught should be part of the patient record so that all healthcare providers can reinforce the teaching points. Several guidelines stressed that patients must understand the importance of communicating worsening or unrelieved pain and reporting adverse effects. Four documents [23, 24, 31, 32] contained patient-specific information on intraspinal drug delivery systems.

Patient safety

Six documents [23, 24, 3133, 35] provided advice on patient safety. Two safety measures were emphasized across all documents: (1) The maintenance of strict aseptic conditions in all aspects of intraspinal analgesia administration, including wearing a mask and gloves when accessing an intraspinal device and (2) the clear labeling of all equipment. All intraspinal drug delivery equipment must be appropriately labeled (yellow for epidural and blue for intrathecal) as to its purpose and the date and time of initiation of infusion.

The use of intraspinal analgesia can confer increased mobility for patients, thus, raising special safety concerns that patients should be aware of such as avoiding scanners in airports and shops, saunas and sunbeds, and sports or activities that may cause injury or dislodgement of the pump. Some intraspinal drug delivery systems are at risk for significant damage and malfunction from magnetic resonance imaging scanners and advice on how to proceed will be required from individual scanning departments and the specific drug delivery system manufacturers. It was advised that patients have a MedicAlert emblem that alerts healthcare professionals to the presence of an intraspinal device in emergency situations or carry an identification card indicating the make and model of any intraspinal device, the drugs within the pump, and the current or last prescribed dose.

Discussion

In general, moderate to severe cancer-related pain tends to be poorly managed. Common barriers to adequate pain control include the belief that patients are not good judges of the severity of their pain; the inability to distinguish between tolerance, physical dependence, and addiction; fear of opioid side effects or addiction; and lack of knowledge about the multiple approaches to managing pain [28]. Because few cancer patients will require intraspinal analgesia, such barriers are magnified with respect to the use of intraspinal analgesia. According to Stearns [9], a general lack of understanding exists about the relevant indications for intraspinal analgesia and the types of patients who would benefit from it. Furthermore, at centers where expertise in intraspinal analgesia is limited, oncologists are unlikely to be familiar with its use and unlikely to refer patients for this type of pain control.

The main advantages of intraspinal analgesia for cancer pain are the delivery of adequate pain control and fewer side effects than with conventional analgesia routes. Disadvantages of intraspinal analgesia include the technically demanding insertion procedure and close patient follow-up required by skilled healthcare personnel.

Because intraspinal techniques are an infrequently used intervention for treating cancer pain, high-quality evidence evaluating their effectiveness and the provision of practical guidance for their delivery are sparse. Some of the most methodologically rigorous documents identified had very little information specifically on intraspinal techniques, while relevant guidance was more often obtained from internal use care pathways or nonevidence-based procedure manuals. The evidence for this organizational guideline was based on two systematic reviews, one consensus conferences, 12 RCTs, and 13 advice documents (practice guidelines, standards, and local policies and procedures). The clinical evidence confirmed the effectiveness of intraspinal analgesia in carefully selected patients with intractable cancer pain. Among the RCTs, some limitations with respect to the small numbers of patients both entered and followed-up were apparent. With this in mind and despite the proportionally small number of patients with cancer pain for whom intraspinal analgesia is the appropriate intervention, this review confirms well-designed randomized trials intending to clarify clinical efficacy issues can reach statistical significance. Adequate accrual, however, will require both a firm commitment from investigators and ensuring adequate funding.

The most clinically relevant studies achieving statistical significance involved the use of nonopioids (i.e., ziconotide and clonidine). This suggests “opioid refractory pain” should serve as a focus for future investigations and given the geographic variation in availability for these particular medications, their use in the setting of intraspinal analgesia could serve as a focus for advocacy.

Advice pertaining to organizational guidance was obtained from practice guidelines, standards, and local policy and procedures manuals. With these documents plus the clinical expertise of the working group, we were able to make recommendations regarding specific indications for use of intraspinal analgesia, practice setting requirements for caring for a cancer patient receiving intraspinal analgesia, the roles and responsibilities of the healthcare professionals involved in providing intraspinal care, the training and education required for healthcare professionals and patients and families, and the elements of patient safety.

Conclusions

As part of a comprehensive cancer pain management strategy, intraspinal catheter insertion should be a consideration for appropriately selected patients. For institutions that have the resources available to safely insert and subsequently manage intraspinal infusions, it is essential that the institution develop the necessary policies, procedures, and competencies to support the healthcare professionals involved in the care of these patients.

Conflict of interest

The authors declared no actual or potential conflicts of interest in relation to this report.

Copyright information

© Springer-Verlag 2009