Background, Objectives, and Methods. A few recent reports suggested that spinal cord stimulation (SCS) effectively suppresses chronic abdominal pain. However, there is no consensus on patient selection or technical aspects of SCS for such pain. That is why we conducted national survey and collected 76 case reports. There were six incompletely filled reports, so we analyzed 70 cases.
Results. There were 43 female and 27 male patients. SCS was trialed in an average of 4.7 days (median of 4 days). In most patients, the leads were positioned for the SCS trial with their tips at the level of the T5 vertebral body (26 patients) or T6 vertebral body (15 patients).
Four patients failed SCS trial: their average baseline visual analog scale (VAS) pain score was 7 ± 2.4 cm and did not improve at the conclusion of the trial (6.5 ± 1.9 cm; P = 0.759). Pain relief exceeded 50% in 66 of 70 patients reported. Among those, VAS pain score before the trial averaged 7.9 ± 1.8 cm. During the trial VAS pain scores decreased to 2.45 ± 1.45 cm (P < 0.001). The opioid use decreased from 128 ± 159 mg of morphine sulfate equivalents a day to 79 ± 112 mg (P < 0.017).
During permanent implantation most of the physicians used two octrode leads and were positioned midline at T5-6 levels. The average patient follow-up was 84 weeks. VAS pain scores before an implant were 8 ± 1.9 cm, while after the implant 2.49 ± 1.9 cm. The opioid use before an implant was 158 ± 160 mg and at the last office visit after the implant 36 ± 49 mg.
Conclusion. In conclusion, it seems that the SCS for the treatment of the abdominal visceral pain may provide a positive patient long-term experience, significant improvements in pain scores and a decrease in opioid use.
Each year, 16 million patients in the United States visit their physician with a chief complaint of abdominal pain. Of these patients about 2 million continue to have persistent severe abdominal pain despite pharmacological interventions, minimally invasive techniques, and surgical interventions [1,2]. Pain is the most prevalent symptom in gastroenterology clinics . Chronic abdominal pain is debilitating and has significant effects on the patient's socioeconomic status often generating strong, affective responses . The need for continuous treatment of chronic abdominal pain also causes a significant burden on the health care system as well as on the patient .
Electrical spinal cord stimulation (SCS) has been successfully used for the treatment of neuropathic pain of various etiologies [4,5]. In animals, electrical SCS can suppress visceral response to colon distension . A few recent case reports involving human subjects suggested that SCS effectively suppresses severe, long-standing abdominal/pelvic visceral pain [7–12].
Although a significant number of patients with severe chronic visceral pain will transiently respond to sympathetic blocks and radiofrequency ablation, substantial long-term pain relief is usually lacking [13,14]. SCS therefore may be a welcome addition to the treatment of moderate to severe chronic visceral abdominal pain. Recently, we reported pain relief, improvement in functional capacity, and decreased consumption of opioids in 30 patients who underwent SCS implantation for control of chronic epigastric visceral pain  and six patients with pelvic visceral pain . Although, there are compelling preliminary data that SCS may decrease pain and improve functional capacity in patients with various visceral chronic pain syndromes, it may require years to accumulate sufficient supportive clinical evidence.
Previously reported indications for the SCS to treat abdominal pain included mesenteric ischemia, irritable bowel syndrome , chronic pancreatitis [9–11], diffuse abdominal adhesions, painful attacks of familial Mediterranean fever , and chronic visceral pelvic pain after long-standing endometriosis . Currently, pain physicians are using SCS in clinical practice for the treatment of abdominal visceral pain based on their own personal experience in placing epidural electrodes and unpublished experiences of other implanting physicians. There is no consensus on patient selection, appropriate lead placement technique and segmental positioning, number of leads needed, or even on lateralization of the leads (e.g. dorsal column or dorsal root ganglion stimulation). Additionally, individual pain physicians may trial only a handful of patients with chronic abdominal pain, mainly because of the current issues with payers. Thus, a substantial number of completed SCS cases (either successful or not), may be dispersed across the country and amenable to review only through survey.
Therefore, our goal in conducting this survey was to understand physicians' current practices relative to SCS use for managing severe abdominal pain. Specifically, we wanted to distinguish various technical aspects of lead placement and identify which of the abdominal pain syndromes were being treated by SCS.
The names and e-mail addresses of pain management physicians who use SCS were collected from the various societies' directories (American Academy of Pain Medicine and North American Neuromodulation Society). The practicing physicians were then contacted using current e-mail or telephone numbers and asked to participate in the study. Case report forms were available on a protected Internet site to participating physicians for completion. We estimated that although about 2,500 physicians use such technology across the country, only a small percentage have actually tried SCS for the management of visceral abdominal pain.
There have been reported differences in responses between e-mail surveys and written surveys . This may make it difficult to assess whether the observed differences are due to factors such as participant deception or actual differences due to characteristics associated with computer and noncomputer users. Therefore, we also called or mailed an invitation to those who did not respond to e-mail queries.
The questionnaire used was named: Case Report-Spinal Cord Stimulation for Visceral Abdominal Pain and contained the following data: Physicians name, e-mail address, study code number, age, and sex of the patient. Data points queried included: causes of abdominal pain (diagnosis) that was treated using the SCS. Pain characteristics of the abdominal pain syndromes were described, as well as the location of particular pain area (epigastric, periumbilical, or other). Also, previous pain treatments are investigated. Prior treatments and diagnostic blocks administered to patients were queried to determine patient and pain characteristics that promoted attempt of SCS.
To evaluate the technical details of SCS trialing, we inquired whether psychological evaluation for the implantable devices was completed and whether a consensus multidisciplinary committee approved trial of SCS. Thereafter, important details on the lead placement were examined: how many leads were placed, at which vertebral level was the final position of the tip of stimulating lead, which type of leads were used (octapolar vs quadripolar, electrode spacing) and the horizontal orientation of the leads (midline, paramedian, and lateral). Also, the number of trialing days, before deciding if a permanent implant was indicated, was determined. Pain scores and opioid use before and after the trialing were inquired to obtain a general understanding from each of the physician contributors on which configuration of the leads and general positioning was thought to be advantageous both for trialing and implantation purposes. We collected similar data for the full implantation.
To evaluate differences between the various values, Mann–Whitney Rank Sum Test was used. Statistical tests and graphs were produced using Sigma Plot software (Systat Software Inc., San Jose, CA). Results are presented as means ± SDs unless otherwise specified.
Seventy-six case reports from 23 responding interventional pain medicine physicians were collected. Six incompletely filled reports were excluded from the analysis, resulting in reporting on 70 cases.
The mean age of the patients who received SCS for chronic abdominal pain was 47.3 years (median 47 years). The age range was from 16 to 85 years of age. There were 43 female and 27 male patients. The most common, identifiable causes of the abdominal pain for which SCS was used were chronic pancreatitis (23 patients) and post-surgical intra-abdominal adhesions (20 patients), followed by the gastroparesis (nine patients). Frequently, post-surgical visceral pain included persistent abdominal pain following common abdominal surgeries like cholecystectomy, bowel resection, etc., but also after the gastric bypass, endometriosis-related surgical procedures, and following the Nissen's fundoplication. There were nine patients where the identifiable cause of their pain could not be determined. “Burning” and “aching” were the most commonly described characteristics of the chronic pain while “throbbing,” “stabbing,” “cramping,” “dull,” and “sharp” were described less frequently. The most frequent painful areas were first, epigastric and second, periumbilical.
Prior to SCS, most of the patients were treated with opoids and a smaller number with membrane stabilizers and/or antidepressants. Among the prior interventional procedures, sympathetic blocks prevailed. Additional clinical interventions in our sample included patients who had pancreatic resection, ileostomy closure, endoscopic retrograde cholangiopancreatography, gastric pacemakers, and multiple laparoscopic adhesiolysis. Only 53% actually received a diagnostic blocks (either celiac or splanchnic) to characterize their visceral pain.
SCS Trials were conducted as follows, 62 out of 70 reported patients received psychological evaluation for the implantable devices just before the trial. Only 24 out of 70 cases (34%) were evaluated by a local multidisciplinary committee for appropriateness of implantable devices before the decision was made to proceed with the trial. The majority of leads were placed in the midline position (Figure 1); however, in 21 cases leads were placed paramedially (P = 0.99). Fifty percent of the patients (35 out of 70) received two leads (mainly octrodes) for their trial. There was no difference (by the end of the trial) in percentage of the pain score improvements if one or two leads were used (P = 0.11). SCS was trialed on average for 4.7 days (median of 4 days) with the shortest trial duration of only 1 day and longest of 14 days. In most patients, the leads were positioned for the SCS trial with their tips at the level of the T5 vertebral body (26 patients, Figure 2) or T6 vertebral body (15 patients). All physicians reported appropriate coverage of the painful area with paresthesias during what was considered above the perception threshold stimulation.
Four patients were reported to have failed SCS trial: their average baseline visual analog scale (VAS) pain score was 7 ± 2.4 cm and did not improve at the conclusion of the trial (6.5 ± 1.9 cm; P = 0.759; Figure 3). They were patients of four different reporting physicians and three patients actually responded to the diagnostic visceral block before the trial.
Pain relief exceeded 50% in 66 of 70 patients reported (94%). Among the 66 patients in whom the trial was deemed successful, VAS pain scores before the trial averaged 7.9 ± 1.8 cm. During the trial, VAS pain scores decreased to 2.45 ± 1.45 cm (P < 0.001; Figure 3). A significant decrease of opioid use from baseline (visit before the trial) was noted during the trial period, but its magnitude was not as profound as that noted when baseline opioid use was compared with the last patient visit after the implantation. In the latter case, opioid use decreased from 128 ± 159 mg to 79 ± 112 mg of morphine sulfate (MSO4) equivalents a day (P < 0.017; Figure 4).
During permanent implantation, similar to the trial phase, most of the physicians who responded to the survey would use two octapolar leads (36 out of 66) and most of which were positioned midline at T5-6 levels (see Figure 2). The average patient follow-up was 84 weeks (median 62 weeks; range from 9–436 weeks of stimulation). Majority (56 out of 66 patients) were followed for more than a year. VAS pain scores before an implant were in average 8 ± 1.9 cm, while after the implant dropped to 2.49 ± 1.9 cm. This improvement in pain scores was significant (P < 0.001; Figure 3). The opioid use before an implant when expressed in MSO4 equivalents was 158 ± 160 mg (median 80 mg) and at the last office visit after the implant 36 ± 49 mg (median 20 mg). Not unexpectedly, the long-term decrease in opioid use noted at the last patient visit was greater than the opioid decrease noted during the trial period (Figure 4). There were eight spinal cord stimulator revisions, two systems were removed because of patient dissatisfaction (one had diarrhea with SCS and the other headache). Overall patient satisfaction as reported by the treating physicians was considered very high (Figure 5). Sixty-two out of 70 patients (89%) were perceived by their providing physicians to have a positive opinion regarding such treatment and 35 patients (50%) were extremely satisfied (Figure 5).
This survey was conducted to describe the technical and clinical variability among interventional pain medicine physicians using SCS for managing abdominal pain across the United States. Assessed were the most frequently used techniques for the patient selection, trialing, and implantation when SCS was attempted for the treatment of visceral abdominal pain. This article can be considered as the preliminary technical report on specifics of this application of neuromodulation.
Considering the number of SCS cases reported by the physicians responding to this survey, SCS for abdominal visceral pain is still rarely used despite its high therapeutic success rate. Reasons may be very few studies describing basic mechanisms of neuromodulation for long-standing visceral pain, comfort levels by the physicians, and issues with coverage of such treatment by the payors. Costs related to SCS include initial implantation costs (from approximately U.S. $20,000–50,000), annual maintenance costs, and complication costs. Approximately 18% per year should be added to the initial implantation costs to cover the costs associated with annual maintenance and complications . It is important to note that despite these costs, there is evidence that initial implantation costs of SCS are offset by the reduction in various post-implant health care costs . The currently available systems are not labeled by the Food and Drug Administration for the specific complaint of abdominal pain, but SCS is approved for stimulation of the trunk. The cases reported in this survey suggest a great potential for this modality of care.
SCS use for abdominal pain is a nascent application. It has been about 6 years from the time when Greeweld van Meerved and others  reported on the suppression of viscero-motor reflex using dorsal column stimulation in rats and about 4 years from the time when the first case reports emerged describing SCS use for managing visceral abdominal pain in humans [8,9]. Since that time, a number of case-reports and case-series were published, all suggesting positive effects of the SCS on pain scores, opioid use, and functional capacity in patients with chronic abdominal pain of various causes [7–12]. Recently, we completed our single-center retrospective study analyzing the trialing successes and long-term improvements in pain scores and the opioid use in 35 consecutive patients who received SCS for various causes of visceral abdominal pain . Those patients were selected based on the results of diagnostic differential epidural nerve block and sympathetic (splanchnic or celiac) nerve blockade. In addition, they underwent psychological evaluation for the implantable devices and were cleared by our Interdisciplinary Committee for the Implantable Devices. A high success rate was noted for trialing of SCS when used for treatment of chronic visceral abdominal pain (86% of the patients had a positive trial and went on to the SCS implantation) and positive long-term responses found in our retrospective study  prompted us to examine the nationwide use of SCS systems for the treatment of intractable abdominal visceral pain. The technical aspects used to achieve the optimal spinal cord stimulation in this surveys patient group (70 patients) are consistent with the data that we collected in our retrospective study on 35 consecutive patients . The most frequent placement of the lead was posterior epidural midline position and the most frequent vertebral level reached with the lead tip was T5 and T6. Sixty-three percent of the patients in our retrospective study  and 56% of the patients in this survey had leads positioned at that vertebral height to achieve optimal paresthesias within the abdominal area and just below the chest.
The proportion of the patients in this survey who received successful trial and went on to have an implant was 94%. Such success rate is very high when compared with success rates of our retrospective study of 35 consecutive patients trialed with SCS for the abdominal, visceral pain (86%). This may imply that nonvisceral abdominal pain may respond better to SCS or more likely that trial failures were under-reported in this survey, suggesting a bias due to pain physicians only participating in the survey if they had positive responses.
Thus, a weakness of this survey is uncertainty as to whether the sample of physicians is representative of physicians using SCS for chronic visceral abdominal pain. To limit selection bias, we attempted to survey the entire group of physicians who might potentially use SCS for visceral abdominal pain, not just those known to use SCS for the treatment of visceral abdominal pain.
To further reduce the effect of response bias, we chose only fully answered survey forms for analysis. Accuracy of the answers lies with the interviewed physicians; however, biased answers are unlikely as data were copied from the patient charts and questions that could not be answered by reviewing the chart data or subjective answers were avoided. Finally, the standard for the success of any survey is how well the data measures the population being studied. In this case, we are surveying pain physicians as surrogates for their patients and asking for outcome information. We purposefully designed the survey to provide information that would be discernable from the patient's medical record. The problem inherent in this analysis, however, is relying upon chart reviews by multiple individuals who are not trained to reliability nor independent observers of treatment outcomes.
In conclusion, it appears from this preliminary snap-shot of SCS use in chronic visceral abdominal pain that the technical aspects of this use of SCS are uniform among the physicians who use such technology around the country. This result is also consistent with the previous larger retrospective study , and also consistent with series of previously published case reports and case-series regarding the lead positioning, choice of leads, days of trialing, etc. Also, we found that interventional pain physicians using such technology reported largely positive long-term outcomes including significant improvements in patient pain scores and decreased opioid use. Although reported follow-up interval varied greatly (from 9 to 436 weeks), most of the patients were followed for longer than 1 year (56 of 70 patients) suggesting prolonged effects of the SCS.
We thank the following physicians for submitting their SCS cases: Drs. Marshall Bedder, Salim Ghazi, Theresa Dews, Todd Jaffe, Prathima Reddy, Silviu Brill, David Walega, Allen W. Burton, Carl Balog, Ross Dickstein, Oscar Deleon Casasola, Padma Gulur, Gladstone C. McDowell, Richard S. Epter, and Sudhir Diwan.