www.kidney-international.org c l i n i ca l i nves t iga t ion

Novel treatment protocol for amelioratingrefractory, chronic pain in patients with autosomaldominant polycystic kidney disease

Niek F. Casteleijn1, Maatje D.A. van Gastel1, Peter J. Blankestijn2, Joost. P.H. Drenth3, Rosa L. de Jager2,Anna M. Leliveld4, Ruud Stellema5, Andreas P. Wolff5, Gerbrand J. Groen5 and Ron T. Gansevoort1; onbehalf of the DIPAK Consortium

1Department of Nephrology, University Medical Center Groningen, University of Groningen, Groningen, Netherlands; 2Department ofNephrology and Hypertension, University Medical Center Utrecht, Utrecht, Netherlands; 3Department of Gastroenterology andHepatology, Radboud Medical Center Nijmegen, University of Nijmegen, Nijmegen, Netherlands; 4Department of Urology, UniversityMedical Center Groningen, University of Groningen, Groningen, Netherlands; and 5Pain Center, Department of Anesthesiology, UniversityMedical Center Groningen, University of Groningen, Groningen, Netherlands

Autosomal dominant polycystic kidney disease (ADPKD)patients can suffer from chronic pain that can be refractoryto conventional treatment, resulting in a wish fornephrectomy. This study aimed to evaluate the effect of amultidisciplinary treatment protocol with sequential nerveblocks on pain relief in ADPKD patients with refractorychronic pain. As a first step a diagnostic, temporary celiacplexus block with local anesthetics was performed. Ifsubstantial pain relief was obtained, the assumption wasthat pain was relayed via the celiac plexus and majorsplanchnic nerves. When pain recurred, patients were thenscheduled for a major splanchnic nerve block withradiofrequency ablation. In cases with no pain relief, it wasassumed that pain was relayed via the aortico-renal plexus,and catheter-based renal denervation was performed. Sixtypatients were referred, of which 44 were eligible. In 36patients the diagnostic celiac plexus block resulted insubstantial pain relief with a change in the median visualanalogue scale (VAS) score pre-post intervention of 50/100.Of these patients, 23 received a major splanchnic nerveblock because pain recurred, with a change in median VASpre-post block of 53/100. In 8 patients without pain reliefafter the diagnostic block, renal denervation wasperformed in 5, with a borderline significant change in themedian VAS pre-post intervention of 20/100. After amedian follow-up of 12 months, 81.8% of the patientsexperienced a sustained improvement in pain intensity,indicating that our treatment protocol is effective inobtaining pain relief in ADPKD patients with refractorychronic pain.Kidney International (2017) -, -–-; http://dx.doi.org/10.1016/

j.kint.2016.12.007

Correspondence: Ron T. Gansevoort, Expertise Center for Polycystic KidneyDiseases, Department of Nephrology, University Medical Center Groningen,P.O. Box 30.001, 9700 RB Groningen, Netherlands. E-mail:R.T.Gansevoort@umcg.nl

Received 10 October 2016; revised 14 November 2016; accepted 1December 2016

Kidney International (2017) -, -–-

KEYWORDS: ADPKD; celiac plexus block; major splanchnic nerve block; pain;

polycystic kidney disease; renal denervation

Copyright ª 2016, International Society of Nephrology. Published by

Elsevier Inc. All rights reserved.

A utosomal dominant polycystic kidney disease(ADPKD) is the most common hereditary renal disor-der, with a prevalence of 4.4 per 10.000.1 In affected

patients numerous cysts are formed in both kidneys and oftenalso in the liver, leading to organ enlargement that can bemassive. Renal function decline is the main clinical problem,leading to end-stage renal disease between the fourth and sev-enth decade of life in most patients.2 Chronic pain is anotherdebilitating complication, with an estimated prevalence of60%.3,4 In a number of cases it can be severe and have a largeimpact on physical and social activity.3,4 In case of pain causedby pressure of the enlarged organs on adjacent tissues or bydistension of the hepatic capsule, pain stimuli are considered tobe relayed via the celiac plexus and major splanchnic nerves,whereas in pain caused by distension of the renal capsule, thepredominant pathway is via the aorticorenal plexus and minorand least splanchnic nerves5 (Figure 1). Chronic pain can bedifficult to manage, and may lead to a need for major analgesictherapy and surgical procedures, such as cyst aspiration, cystfenestration, or even nephrectomy.6,7 In the literature it hasbeen suggested that nerve blocks can be used for pain man-agement before such invasive therapies are explored.6–9 How-ever, no study has been performed to investigate the effect ofnerve blocks on pain relief in ADPKD patients.

We recently proposed a novel approach for treatment ofrefractory chronic pain in ADPKD that applies sequentialnerve blocks.8 When after a multidisciplinary assessmentnon-ADPKD–related causes are ruled out, a diagnostic,temporary celiac plexus block with a local anesthetic agent isperformed. In cases in which substantial pain relief is ob-tained, it is assumed that pain was caused by pressure onadjacent tissues or distension of the hepatic capsule. Conse-quently, when pain recurs, a long-term block of the major

1

Th5

Th6

Th7

Th8

Th9

Th10

Th12

Th11

Upper abdominal

organs (incl. liver)

Spinalcord

L1

ST

Aorta

Celiac plexus

Aorticorenal plexus

1.

2.

3.

1. Celiac plexus2. Major splanchnic

nerves3. Perivascular

renal denervation

Proposed block

Figure 1 | Schematic representation of the sensory nerve supplyof the kidneys and upper abdominal organs. Solid line: majorsplanchnic nerve providing sensory innervation of the upperabdominal organs, including the liver via the celiac plexus. Dottedline: lesser splanchnic nerve providing sensory innervation of therenal parenchyma and ureter. Dashed line: least splanchnic nerveproviding sensory innervation of the renal capsule. The perivascularnerve plexus around the renal artery forms the final commonpathway to and from the kidney. ST, sympathetic trunk. Adapted withpermission from Bajwa ZH, Gupta S, Warfield CA, et al. Painmanagement in polycystic kidney disease. Kidney Int. 2001;60:1631–1644.6

c l i n i ca l i nves t iga t i on NF Casteleijn et al.: Chronic pain management in ADPKD

splanchnic nerves by radiofrequency ablation (RF-MSNblock) is performed. When there is no response to the diag-nostic celiac plexus block, pain stimuli are likely to be relayedvia the alternative pathway (i.e., the aorticorenal plexus), inwhich case renal denervation is the preferred option. Thisintervention is executed via a catheter-based technique,originally developed as treatment for refractory hypertension,

2

and has recently been suggested as an effective treatment ofchronic pain in selected ADPKD patients.10,11 Here we pre-sent the results of our multidisciplinary protocol consisting ofsequential nerve blocks in ADPKD patients with refractorychronic pain.

RESULTS

Patient characteristicsA total of 60 patients visited our Expertise Center for Poly-cystic Kidney Diseases for analysis and treatment of refractorychronic ADPKD-related pain, of which 3 patients were self-referred. After assessment of inclusion and exclusioncriteria, 44 patients were deemed eligible to participate in ourtreatment protocol (Figure 2). Sixteen patients were ineligiblebecause another treatment option was chosen (such as ne-phrectomy in patients undergoing renal replacement therapy[RRT) or cyst aspiration in case of a limited number of verylarge cysts], pain was likely non-ADPKD–related, pain couldbe treated with additional medication, or the patients rejectedthe treatment protocol. Characteristics of these ineligiblepatients are given in Supplementary Tables S1 and S2. Meanage of the included patients was 50 � 9 years, and 77.3% werefemale (Table 1). Three patients were RRT-dependent, and inthe non-RRT–dependent patients (n ¼ 41), mean estimatedglomerular filtration rate (eGFR) was 57 � 25 ml/min/1.73m2. Pain was present for a median period of 7 years(interquartile range [IQR]: 4–18) and was experienced asrefractory for 12 months (IQR: 10–24) (Table 2). Nearly allpatients (95.5%) used daily opioids, except one who had acontraindication against opioid use, and 18 (40.9%) hadpreviously been treated by invasive therapies such as cystaspiration (n ¼ 8), cyst sclerotherapy (n ¼ 4), cyst fenestra-tion (n ¼ 5), or contralateral nephrectomy (n ¼ 5). Pain hadan impact on patient’s work, daily activities, and social life, asindicated by low physical component score (PCS) and mentalcomponent score (MCS) (34 � 17 and 50 � 21, respectively).No associations of total kidney volume, total liver volume, orcombined kidney and liver volume with visual analogue scale(VAS) score were found (P ¼ 0.6, P ¼ 0.3, and P ¼ 0.5,respectively).

Sequential blocksIn all 44 patients an ipsilateral, diagnostic, temporary celiacplexus block with 10 ml of bupivacaine 0.5% was performed.Kidney size was no limitation to perform a nerve block. In 36patients (81.8%), substantial pain relief was obtained (medianchange in VAS pre-post intervention: 50/100 [IQR: 26–68];P < 0.001) (Figure 2 and Table 3). In 13 patients (36.1%) paindid not recur (i.e., remained below 50/100) after the initialceliac plexus block and no further action was taken (medianchange in VAS pre-post intervention: 60/100 [IQR: 35–70];P < 0.002) (Table 3). Twelve of these 13 patients were nolonger dependent on daily-use opioids, and only 5 used dailyacetaminophen.

In the remaining 23 patients (63.9%), pain recurred after amedian follow-up of 6 (IQR: 3–11] weeks, for which reason

Kidney International (2017) -, -–-

YES

RF-MSN block: 23 patients

NO

No protocolized treatment in 16 patients:- Option rejected by patient: 4- Pain not invalidating: 5- Pain deemed not ADPKD-related: 2- Cyst aspiration as preferred therapy: 2- Nephrectomy as preferred therapy: 3

No further action:20 patients

Positive block:36 patients

Renal denervation:8 patients

Screened:60 patients

No further action:13 patients

Sustained pain relief?

Sustained pain relief?

NO

NO

YES

YES

No further action:1 patient

Additonal treatment:1 patient

No further action:3 patients

Sustained pain relief? YES

NO

Substantial pain relief?

Diagnostic celiac block:44 patients

Option rejected:3 patients

No further action:2 patient

Additonal treatment:1 patient

Figure 2 | Flowchart of patients screened for treatment according to the multidisciplinary protocol for refractory chronic pain inautosomal dominant polycystic kidney disease.

NF Casteleijn et al.: Chronic pain management in ADPKD c l i n i ca l i nves t iga t ion

the patients were scheduled for an RF-MSN block. In 2 pa-tients the ipsilateral RF-MSN block was performed twice dueto initial moderate success and ultimately resulted in sub-stantial pain relief in both patients. A bilateral RF-MSN blockwas performed in 5 (13.9%) patients, because they experi-enced left- as well as right-sided pain. Ultimately, the medianchange in VAS pre-post RF-MSN was 53/100 (IQR: 23–65],measured at 2 to 4 weeks after the final procedure (P <0.001). In 20 of the 23 patients (87.0%), a substantial andsustained improvement in pain intensity was observed,leading to cessation of daily opioid use in 16 patients (69.6%)

Kidney International (2017) -, -–-

and a decrease in dosage in 4 (17.4%) (Table 3). Of the 3patients without pain relief after the RF-MSN block, addi-tional treatment was given to 1. A diagnostic, temporarysympathetic block with local anesthetics was applied at thelevel of L2 with success, but the subsequent long-term RF-MSN block resulted in a decrease in pain of only 20%.

The 8 patients (18.2%) without a response to the initialceliac plexus block were scheduled for catheter-based renaldenervation. This procedure was performed in 5, because 3patients rejected this option. The median change in VAS pre-post renal denervation was 20/100 (IQR: 0–50), measured at

3

Table 1 | Patient characteristics (N [ 44)

Age (yr) 50 � 9Female sex (%) 77.3Height (cm) 173 � 8Weight (kg) 80 � 16Body mass index (kg/m2) 27 � 4History of- Urinary tract infection (%) 65.9- Renal cyst infection (%) 22.7- Liver cyst infection (%) 6.8- Bouts of macroscopic hematuria (%) 59.1- Renal stones (%) 13.6- Renal surgery (%) 18.2- Liver surgery (%) 6.8- Liver cysts (%) 95.3Systolic blood pressure (mm Hg) 132 � 12Diastolic blood pressure (mm Hg) 84 � 8Use of blood pressure–lowering drugs (%) 75.0Non-RRT dependent (%) 93.2- eGFR (ml/min/1.73 m2) 57 � 25Renal transplantation (%) 6.8- eGFR (ml/min/1.73 m2) 52 � 14Short Form-36 Score- Physical component score, 0–100) 34 � 17- Mental component score, 0–100) 50 � 21Organ volumes (median [IQR])- Left kidney (ml) 874 (548–1309)- Right kidney (ml) 854 (545–1326)- Total kidney (ml) 1664 (932–2609)- Liver (ml) 2612 (1944–3327)- Total kidney and liver (ml) 4446 (3427–5695)

eGFR, estimated glomerular filtration rate; IQR, interquartile range; min, minute; RRT,renal replacement therapy; yr, years.

Table 2 | Pain characteristics (N [ 44)

Duration of- Pain (yr) 7 [4–18]- Refractory pain (mo) 12 [10–24]Pain severity last 4 weeks- Minimum VAS score, 0–100 40 [21–60]- Maximum VAS score, 0–100 88 [80–90]- Average VAS score, 0–100 70 [55–80]Patient reported location as- Left kidney (%) 65.9- Ventral side 27.6- Dorsal side 72.4

- Right kidney (%) 52.2- Ventral side 39.1- Dorsal side 60.9

- Liver (%) 27.2- Ventral side 66.7- Dorsal side 33.3

Management of pain- Nonpharmacologic therapies (%) 65.9- Heat pads 45.5- Physiotherapy 40.9- Cognitive behavioral therapy 11.4

- Pharmacologic therapies (%) 100.0- Acetaminophen 74.4- NSAID 2.3- Sleep medication 13.6- Low-dose opioids 45.5- High-dose opioids 50.0

- Previous invasive pain therapies (%) 40.9- Nerve block 6.8- Cyst aspiration 18.2- Cyst sclerotherapy 9.1- Cyst fenestration 11.4- (Partial) nephrectomy 11.4

mo, months; NSAID, nonsteroidal anti-inflammatory drug; VAS, visual analogue scale;yr, years.

c l i n i ca l i nves t iga t i on NF Casteleijn et al.: Chronic pain management in ADPKD

2 to 4 weeks after the procedure (P ¼ 0.07). In 3 patients asustained improvement in pain intensity was observed,leading to cessation of daily opioid use. In the remaining 2patients no pain relief was noticed. Additional treatment (i.e.,diagnostic sympathetic block with local anesthetics at the levelof L2) was given in 1 of these 2 patients with success, but thesubsequent long-term RF block did not lead to a decrease inpain.

In the overall group of included patients we observed anincrease in quality of life (change in PCS: þ7 [IQR: 0–20],P ¼ 0.001; change in MCS: þ5 [IQR: –1 to 13], P ¼ 0.04). Ofnote, a formal statistical significant effect was not reached inall subgroups on these quality of life measures, likely due tothe small number of patients in some subgroups (e.g., therenal denervation group; Table 3). Characteristics of indi-vidual patients in the various aforementioned subgroupsare presented in supplementary files (SupplementaryTables S3–S8).

Follow-upAfter a follow-up of 12 (IQR: 8–17) months, 81.8% of the 44patients that underwent 1 or more nerve blocks experienced asustained improvement in pain intensity (median change inVAS pre-post intervention: 53 [IQR: 35–70], P < 0.001).Daily opioid use was stopped in 63.6% of the patients. Aconsiderable number of cases continued to have intermittentabdominal discomfort (VAS score: 20 [IQR: 20–30]), which

4

in general could be managed adequately with on-demandacetaminophen (54.5%). These results were similar in pa-tients with a follow-up longer than 18 months (n ¼ 9), with77.8% reporting a sustained improvement in pain (medianchange in VAS pre-post intervention: 55 [IQR: 38–73],P < 0.001) and only 2 of 9 patients still using daily opioids,but at a lower dose.

Effect on blood pressure and renal functionIn patients, who underwent a celiac plexus and/or RF-MSNblock, a decrease in blood pressure was observed (medianchange in systolic blood pressure: –5 [IQR: –9 to þ2]mm Hg, P ¼ 0.1; median change in diastolic blood pressure:–4 [–8 to 0] mm Hg, P ¼ 0.01) (Table 3). None of thesepatients had a change in type or dosage of blood pressure–lowering drugs. In the renal denervation group, a similar ef-fect on the blood pressure was seen (median change in sys-tolic blood pressure: 0 [IQR: –17 to þ3] mm Hg, P ¼ 0.1;median change in diastolic blood pressure: –3 [–7 to 0]mm Hg, P ¼ 0.01, respectively), but this effect was obtainedwhile in 80.0% patients the dosage of antihypertensivetreatment was reduced. The procedures had no influence onkidney function (eGFR preintervention: 56 � 24 ml/min/1.73m2; postintervention: 52 � 24 ml/min/1.73m2, P ¼ 0.3).

Kidney International (2017) -, -–-

Table 3 | Results of the diagnostic celiac plexus block, RF-MSNblock, and renal denervation

Intervention

P valuePre Post

Overall (N [ 44)Substantial pain relief (n [%]) X 36 (81.8)VAS score, 0-100 (median [IQR]) 70 (55–80) 18 (0–30) <0.001DDD nonopioids 0.9 � 0.6 0.5 � 0.6 0.003DDD opioids 0.3 � 0.4 0.1 � 0.2 0.06Physical component score, 0–100 34 � 17 44 � 19 0.001Mental component score, 0–100 50 � 21 55 � 23 0.04SBP (mm Hg) 132 � 12 128 � 12 0.1DBP (mm Hg) 84 � 8 81 � 8 0.01DDD BPLD 1.8 � 0.3 1.8 � 0.3 0.3eGFR (ml/min/1.73 m2) 56 � 24 52 � 24 0.3Follow-up (mo) X 12 (8–17)

Diagnostic celiac plexus block (n [ 13)

Substantial pain relief (%) X 13 (100.0)VAS score, 0–100 70 (50–80) 10 (0–28) 0.002DDD nonopioids 0.9 � 0.6 0.3 � 0.6 0.1DDD opioids 0.3 � 0.4 0.02 � 0.01 0.03Physical component score, 0-100 37 � 20 44 � 20 0.6Mental component score, 0-100 51 � 22 59 � 22 0.3SBP (mm Hg) 131 � 10 132 � 17 0.3DBP (mm Hg) 80 � 8 79 � 8 0.3DDD BPLD 1.2 � 0.5 1.2 � 0.5 1.0eGFR (ml/min/1.73 m2) 58 � 31 53 � 33 0.3Follow-up (mo) X 11 (6–15)

RF-MSN block (n [ 23)

Substantial pain relief (%) X 20 (87.0)VAS score, 0–100 70 (60–80) 13 (0–28) <0.001DDD nonopioids 0.8 � 0.7 0.5 � 0.6 0.02DDD opioids 0.3 � 0.4 0.1 � 0.2 0.1Physical component score, 0–100 33 � 16 45 � 20 0.001Mental component score, 0–100 48 � 21 54 � 22 0.07SBP (mm Hg) 132 � 15 127 � 10 0.2DBP (mm Hg) 81 � 9 81 � 9 0.1DDD BPLD 1.7 � 0.9 1.7 � 0.9 1.0eGFR (ml/min/1.73 m2) 60 � 20 55 � 19 0.6Follow-up (mo) X 11 (7–14)

Catheter-based renal denervation (n [ 5)

Substantial pain relief (%) X 3 (60.0)VAS score, 0–100 60 (50–75) 20 (0–20) 0.07DDD nonopioids 0.8 � 0.6 0.5 � 0.6 0.4DDD opioids 0.4 � 0.7 0.2 � 0.3 0.7Physical component score, 0–100 43 � 18 44 � 11 0.8Mental component score, 0–100 53 � 30 63 � 33 0.1SBP (mm Hg) 134 � 6 126 � 9 0.01DBP (mm Hg) 86 � 6 83 � 9 0.2DDD BPLD 2.1 � 0.9 2.0 � 0.9 0.3eGFR (ml/min/1.73 m2) 40 � 27 39 � 28 0.3Follow-up (mo) X 15 (12–19)

BPLD, blood pressure–lowering drug; DBP, diastolic blood pressure; DDD, defineddaily dose; eGFR, estimated glomerular filtration rate; IQR, interquartile range; mo,months; RF-MSN block, radiofrequency ablation block of major splanchnic nerves;SBP, systolic blood pressure; VAS, visual analogue scale.In all patients (N ¼ 44) a diagnostic celiac plexus block was performed. No furtheraction was taken when pain relief was obtained (n ¼ 13). When pain recurred afterthe celiac plexus block, an RF-MSN block was performed (n ¼ 23). When no painrelief after the diagnostic celiac plexus block was obtained, renal denervation wasperformed (n ¼ 8). Two to 4 weeks after the interventions, pain score, quality of life,blood pressure, and renal function were monitored.

NF Casteleijn et al.: Chronic pain management in ADPKD c l i n i ca l i nves t iga t ion

Kidney International (2017) -, -–-

Adverse eventsTable 4 summarizes all adverse events related to the sequentialnerve blocks. Two patients experienced orthostatic hypoten-sion immediately after the diagnostic celiac plexus block,which was self-limiting within 4 hours. Another patient re-ported diarrhea after this intervention, which stopped within3 days without the need for additional treatment. In 1 patientblood was aspirated during the diagnostic celiac plexus blockprocedure, resulting in the procedure being interrupted thenrepeated after 4 weeks with success. In 3 patients the RF-MSNblock was extremely painful, resulting in the procedure beingbriefly interrupted but successfully completed. Renal dener-vation was technically successful in 4 of 5 patients. In the fifthpatient, spasms in the left renal artery occurred when thecatheter was introduced. The procedure was interrupted andsuccessfully repeated after 3 months.

Two patients were hospitalized during the first 8 weeksafter the RF-MSN block. One patient developed dyspnea withfever 7 weeks after the intervention and was admitted to thehospital for 1 day, with hyperventilation diagnosed. Anotherpatient reported an episode of a cerebrovascular transientischemic attack 4 weeks after the RF-MSN block and washospitalized for 2 days. Both incidents were judged to be notrelated to the RF-MSN procedure.

DISCUSSIONThese data show that our treatment protocol, applyingsequential nerve blocks, results in substantial pain relief inADPKD patients with refractory chronic pain. After a follow-up of 12 months, the majority of eligible patients experienceda sustained improvement in pain intensity. Furthermore, weobserved an increase in quality of life. No procedure-relatedserious adverse events or decrease in eGFR were noticed.

At present, no study has been performed to systematicallyinvestigate the effect of nerve blocks, including celiac plexusblocks, on pain in ADPKD patients. Several studies analyzedthe effect of cyst aspiration, sclerotherapy, and fenestration astreatment options for chronic pain in ADPKD patients whennonpharmacological and pharmacological options fail.12–15

The success rates of these interventions were highly variable(i.e., between 20% and 80%). Given the uncertain success rateand risk for complications such as infection, these treatmentoptions are not widely performed. Other, more invasivetreatment options include surgical nephrectomy and trans-cutaneous arterial embolization. According to the literature,these options can result in adequate pain relief, but both leadinherently to a decrease in renal function.6,7,9,16 Because renalfunction decline, with ultimate need for renal replacementtherapy, is the main clinical problem in ADPKD, there is aneed for kidney function–sparing techniques. The presentdata provide evidence that sequential nerve blocks should beconsidered and tried before more invasive therapies areexplored.

5

Table 4 | All adverse events related to sequential nerve blocks(N [ 44)

AllLikelyrelated

Remotelyrelated

Notrelated

During procedure (%)- Blood aspiration during procedure 2.3 2.3- Spasm renal artery 2.3 2.3- Severe pain that procedure wasinterrupted

6.8 6.8

Immediately postprocedure (%)- Pain located on the needleposition

56.8 56.8

- Orthostatic hypotension 4.5 4.5- Diarrhea 2.3 2.3Hospitalization during first 8 weeksafter procedure (%)

- Hyperventilation 2.3 2.3- Cerebrovascular transientischemic attack

2.3 2.3

c l i n i ca l i nves t iga t i on NF Casteleijn et al.: Chronic pain management in ADPKD

Celiac plexus and RF-MSN blocks have proven to beeffective in treatment of refractory chronic abdominal painrelated to, for instance, chronic pancreatitis and pancreatic,gastric, and intestinal cancer.17–19 Renal denervation is nowmainly applied in patients with therapy-resistant hyperten-sion and heart failure,20 but may also be effective for treat-ment of chronic pain syndromes. Two older studies describedrenal denervation for ADPKD-related pain by thoracoscopicand laparoscopic procedures.21,22 In a recent case report, wewere the first to report that catheter-based renal denervationcan also successfully treat pain in patients with ADPKD,10,23

and is a far less invasive procedure. The present study addsevidence in a relatively large series of patients that this pro-cedure is simple, safe, and effective. Catheter-based renaldenervation should, however, only be performed in selectedpatients, because our data indicate that in a minority of pa-tients chronic pain stimuli are relayed via the aorticorenalplexus. The protocol used in this study may aid in the se-lection of ADPKD patients for catheter-based renaldenervation.

It should be noted that we did not perform sequentialnerve blocks in all patients who were referred. When pain wasnot refractory or not ADPKD-related, we first optimizedanalgesic use and treated the other causes. In cases in whichpatients were RRT-dependent, we preferably performed ne-phrectomy, and in cases in which patients had a limitednumber of extremely enlarged cysts, cyst aspiration or cystfenestration was the first-line treatment. These procedures (3nephrectomies and 2 cyst aspirations) led to adequate andsustained pain relief in 4 of the 5 patients in which they wereapplied. In addition, the majority of treated patients werefemale. This is in line with 2 other large studies in ADPKDpatients, in which it was observed that pain was more com-mon in female than male patients.4,24 Whether this sex dif-ference is specific to ADPKD is not clear. Several reviewsconcluded, for instance, that in the general population pain ismore frequently reported by women.25,26

6

In a number of cases with a positive response to thediagnostic temporary celiac plexus block, this interventionresulted in a sustained pain relief, even up to 2.5 years. This issurprising, because local anesthetics are only able to interrupta sensory pathway for a maximum of 24 hours. A possibleexplanation for this unexpected finding may be an effect oncentral sensitization caused by longstanding nociceptivestimulation in the past (e.g., from a cyst infection or cystbleeding). As part of this sensitization process, activationthresholds of sensory neurons decrease and their excitabilityincreases.7,17 Consequently, minor stimuli will lead to a painresponse that normally would not occur. We hypothesize thatby applying local anesthetics, the continuous excitation ofvisceral nociceptive neurons is temporarily interrupted andthe neurons may return to their normal resting potential.17

In some patients no pain relief was obtained or painrecurred. In 2 patients the diagnostic celiac plexus block andsubsequent renal denervation were both unsuccessful. Thismay imply that nociceptive stimuli followed a pathwaydifferent from the ones that were blocked. It has been sug-gested that small sensory nerve connections, which do nottravel via the renal artery, can exist between the renal plexusand the renal capsule.5 These sensory nerve fibers will not beblocked by catheter-based renal denervation. Another expla-nation may be that, as technical failure, not all targeted sen-sory pathways were blocked. For instance, the spiralingablation technique for renal denervation may not havecompletely blocked the aorticorenal pathway. In 3 patients,pain recurred after a positive diagnostic celiac plexus blockand subsequent positive ipsilateral RF-MSN block. In 2 of 3,pain recurred after an acute painful event (cyst bleeding andinfection), which suggests that the RF-MSN block may havebeen incomplete and that remaining sensory nerve fibersrelayed the new nociceptive stimuli to the spinal cord.

When considering sequential nerve blocks for treatment ofchronic ADPKD-related pain in clinical care, the expectedbenefits should of course outweigh potential disadvantages.Common side effects of these specific nerve blocks are severepain during the procedure, pain at the injection site imme-diately after the procedure, or orthostatic hypotension. Theseevents were also reported in 6.8%, 56.8%, and 4.5% of ourpatients, respectively. Two severe adverse events wereobserved, but we deemed these hospitalizations unrelated tothe procedure, because both events were observed more than4 weeks after the intervention and we were not able tocorrelate these events causally to the procedures. However,nerve blocks could have late negative clinical consequences.An RF-MSN block interrupts the upper abdominal sensorynerve supply, leading to a limited or altered nociceptivesensory function in the upper abdomen. Clinicians and pa-tients should be aware that abdominal diseases may thereforepresent with a different symptomatology (i.e., an altered painsensation), which may lead to undesired delay for bothphysician and patient. Of note, we performed this treatmentprotocol only in patients with ADPKD. However, other pa-tients with chronic, refractory kidney pain related to a

Kidney International (2017) -, -–-

NF Casteleijn et al.: Chronic pain management in ADPKD c l i n i ca l i nves t iga t ion

nonmalignant and noninfectious cause, such as loin painhematuria syndrome or symptomatic para-pelvic cysts, mayalso benefit from our novel approach.10,23

This study has limitations, of which the most important isthe nonrandomized, single-center design. We chose toperform this study in such a setting because we considered itunethical to perform sham procedures in patients with re-fractory chronic pain, in line with literature on placeboanesthetic blocks.27 Because not all medical centers haveexpertise with sequential nerve blocks and treatment ofchronic pain in ADPKD patients, and the prevalence of suchpatients is relatively low, treatment was performed in 1 center.The main strength of our study is the systematic and pro-spective nature of data collection, including information onquality of life, that resulted in a well-characterized population.

In conclusion, the present study indicates that our novelmultidisciplinary treatment protocol of applying sequentialnerve blocks is effective in obtaining substantial and sustainedpain relief in ADPKD patients with chronic refractory pain.Patients should be carefully selected for eligibility, and othertreatment options should be considered for ineligible patients.We advise, therefore, that sequential nerve blocks only beperformed in this patient group in a protocolized setting incenters with expertise in treatment of ADPKD-related pain.However, it should be kept in mind that altered pain sensa-tion may lead to a different symptomatology of laterabdominal disease.

MATERIALS AND METHODSStudy populationADPKD patients with refractory chronic pain were screened foreligibility between August 2013 and May 2016 at our ExpertiseCenter for Polycystic Kidney Diseases of the University MedicalCenter Groningen, the Netherlands. Patients were referred by theirtreating nephrologist or were self-referred from all over the coun-try. Patients were eligible if pain was present $3 months; had aseverity on a VAS score of $50 out of 100; limited the patient inwork, daily activities, and social life; and responded insufficiently toopioid treatment, or such treatment was contraindicated. Patientswere excluded when after a multidisciplinary assessment pain wasdeemed not to be ADPKD-related or when invasive therapies (suchas cyst aspiration or nephrectomy) were found to be a better optionto achieve pain relief. The institutional research board concludedthat this protocol was exempted from institutional review boardapproval, because it was considered to be a protocolized intro-duction of novel clinical care (METc 2013.299).

Study assessmentsAll patients were screened by a nephrologist and a pain specialist.Before intake, all patients filled out a questionnaire to rate their painintensity by a VAS score (scale: 0–100) and their quality of life by theShort Form-36 Health Survey. The Short Form-36 scores wereaggregated into a PCS and MCS.28 PCS and MCS were scored from0 to 100, with a higher score reflecting better quality of life. Duringan interview information was collected on demographics, medicalhistory, medication use, pain, and gastrointestinal symptoms. Renalpain was defined as pain or discomfort located in the flank, lowerback, or abdomen. Liver pain was defined as pain or discomfort

Kidney International (2017) -, -–-

located in the right upper abdomen, behind or below the rib cage.Blood pressure was assessed with an automatic device during rest insupine position for 5 minutes. After intake, blood and urine sampleswere collected for routine laboratory testing. Serum creatinine wasused to estimate GFR (applying the CKD-EPI equation).29 All pa-tients underwent magnetic resonance imaging with assessment by aradiologist to exclude other anatomic causes for pain, and formeasurement of total kidney volume and total liver volume. Inaddition, the location of the abdominal aorta and celiac plexus wasidentified, especially to check for potential displacement by theenlarged kidneys and liver. In cases in which the nephrologist andpain specialist agreed that pain appeared to be related to the cysticdisease, patients were scheduled for a diagnostic celiac plexus block.In unclear cases, patients were discussed by a multidisciplinary teamconsisting of a nephrologist, pain specialist, radiologist, urologist,and gastroenterologist, as well as a transplant surgeon and gyne-cologist when needed.

Study proceduresThe diagnostic, temporary celiac plexus block was performed on theside where the patient reported the highest level of pain. The patientwas placed in a prone position with a pillow under the abdomen toreduce lumbar lordosis. A 20-gauge 15-cm spinal-type needle(Cosman Medical, Burlington, MA) was advanced from posterior toanterior toward the ventral surface of the L1 vertebral body underfluoroscopic guidance. After the needle position was confirmed byinjection of contrast medium (to be spread directly anteriorly fromthe L1 vertebral body in lateral fluoroscopic view; and in anterior-posterior view within the bilateral vertebral body borders), 10 mlof bupivacaine (0.5%) was injected (Figure 3). Patients wereobserved closely for 2 to 4 hours after the procedure, including vitalsigns monitoring.

In cases in which substantial pain relief was observed (i.e.,reaching a VAS score #30/100), patients were scheduled for a long-term RF-MSN block when pain recurred with a severity of >50/100.A 20-gauge, 15-cm spinal-type needle (Cosman) was advanced fromposterior to anterior toward the ventral 1/3 surface of the vertebralbody of Th11. Positioning took place under fluoroscopic guidanceand was deemed correct when there was bone contact. After thecorrect needle position was confirmed by injection of contrast me-dium by direct anterior spread to the Th11 vertebral body in lateraland anterior-posterior view, 3 applications of radiofrequency energyat 80�C were executed with 3 mm interspatial space between everyapplication in posterior direction, starting on the first-most anteriorneedle tip position. The time needed to perform a diagnostic celiacplexus block or RF-MSN block was 20 to 30 minutes. Patients wereobserved closely for 2 to 4 hours after the procedure, including vitalsigns monitoring.

In cases in which no substantial pain relief was observed after thediagnostic celiac plexus block, patients were referred to the Uni-versity Medical Center Utrecht for catheter-based renal denervation.Renal denervation of afferent sensory nerves was performed usingthe Simplicity Catheter System (Medtronic, Dublin, Ireland), a 6 Fr–compatible, single-use RF probe. Before introduction of the RFprobe, a renal angiogram was performed to exclude contraindica-tions for the procedure, such as renovascular abnormalities(including renal artery stenosis) and previous renal stent or angio-plasty. The catheter was placed at the most distal location in the renalartery possible, because in the distal segment the sensory nervestravel closer to the arterial lumen compared with the proximal andmiddle segments (Figure 4).30 The catheter was connected to an

7

Th12Left kidney Right kidney

Pancreas Inf. Vena Cava

Aorta

Stomach Liver

Figure 3 | Diagnostic celiac plexus block. Left panel: schematic drawing of a diagnostic celiac plexus block. Right panel: anteroposteriorradiograph showing an example of a diagnostic celiac plexus block near the vertebral column. The solid white line represents the needle.

c l i n i ca l i nves t iga t i on NF Casteleijn et al.: Chronic pain management in ADPKD

automatic RF generator, and applications of RF energy in a spiralpattern along the renal artery, from distal to proximal and with 5mm interspaces, were performed. The time needed to performroutinely catheter-based renal denervation was approximately 45minutes. Patients were observed closely for 24 hours after the

Figure 4 | Renal denervation procedure and peri-arterial renal sensodenervation procedure. The solid white line represents the Simplicity Calocation possible in the renal artery. The dashed line represents the outerperi-arterial renal sensory nerve locations. In the distal segment, the sensituation in the proximal and middle segments. Adapted from Sakakuraarterial renal nerves in man. J Am Coll Cardiol. 2014;64:635–643.30

8

procedure, including vital signs monitoring. During all nerve blocks,patient’s pain experience was not assessed. Diagnostic plexus celiacblock and RF-MSN block were performed under local anesthesia andcatheter-based renal denervation with sedation. This makes it diffi-cult to evaluate during the procedure whether the intervention was

ry nerve locations. Left panel: angiography during the renaltheter System. The catheter electrode is positioned at the most distalborder of the polycystic kidney. Right panel: schematic drawing of thesory nerves travel closer to the arterial lumen when compared to theK, Ladich E, Cheng Q, et al. Anatomic assessment of sympathetic peri-

Kidney International (2017) -, -–-

NF Casteleijn et al.: Chronic pain management in ADPKD c l i n i ca l i nves t iga t ion

successful. In addition, nerve blocks themselves can be painful,which also makes it difficult to assess success of the denervationduring the procedure.

Two to 4 weeks after all interventions, VAS score, defined dailydose of analgesic use, quality of life with the Short Form-36, renalfunction, and blood pressure were monitored. Adverse eventsoccurring during the treatment protocol were recorded.

Statistical analysesNormally distributed variables are expressed as mean � SD, whereasnon-normally distributed variables are given as median (IQR). Dif-ferences in baseline characteristics between eligible and ineligiblepatients were calculated with a chi-square test for categorical data,and for continuous data with Student’s t-test or a Mann-Whitney Utest in case of non-normally distributed data. A paired Student’s t-test or Wilcoxon signed-rank test for non-normally distributed datawas used to compare VAS score, PCS, MCS, blood pressure, defineddaily dose of analgesics, defined daily dose of blood pressure–lowering drugs, and eGFR before and after intervention. Statisticalanalyses were performed using SPSS 22 (SPSS Statistics, Inc., Chi-cago, IL). A 2-tailed P value <0.05 was considered to indicate sta-tistical significance.

DISCLOSUREAll the authors declared no competing interests.

ACKNOWLEDGMENTSDIPAK ConsortiumThe DIPAK Consortium is an inter-university collaboration in theNetherlands that is established to study autosomal dominantpolycystic kidney disease and to develop rational treatment strategiesfor this disease. The DIPAK Consortium is sponsored by the DutchKidney Foundation (grants CP10.12 and CP15.01) and Dutchgovernment (LSHM15018). Principal investigators are (in alphabeticalorder): J.P.H. Drenth (Department of Gastroenterology andHepatology, Radboud University Medical Center Nijmegen,Netherlands), J.W. de Fijter (Department of Nephrology, LeidenUniversity Medical Center, Netherlands), R.T. Gansevoort (Departmentof Nephrology, University Medical Center Groningen, Netherlands),D.J.M. Peters (Department of Human Genetics, Leiden UniversityMedical Center, Netherlands), J. Wetzels (Department of Nephrology,Radboud University Medical Center Nijmegen, Netherlands) and R.Zietse (Department of Internal Medicine, Erasmus Medical CenterRotterdam, Netherlands).

SUPPLEMENTARY MATERIALTable S1. Patient characteristics stratified according to eligibility forthe multidisciplinary treatment protocol consisting of sequentialnerve blocks (N ¼ 60).Table S2. Pain characteristics stratified for eligibility in our treatmentprotocol (N ¼ 60).Table S3A. Characteristics of individual patients responding to atemporary celiac plexus block with sustained pain relief (N ¼ 13).Table S3B. Pain characteristics of individual patients responding to atemporary celiac plexus block with sustained pain relief (N ¼ 13).Table S4A. Characteristics of individual patients responding to along-term major splanchnic nerve block with sustained pain relief(N ¼ 20).Table S4B. Pain characteristics of individual patients responding to along-term major splanchnic nerve block with sustained pain relief(N ¼ 20).

Kidney International (2017) -, -–-

Table S5A. Characteristics of individual patients responding to renaldenervation with sustained pain relief (N ¼ 3).Table S5B. Pain characteristics of individual patients responding torenal denervation with sustained pain relief (N ¼ 3).Table S6A. Characteristics of individual patients with no sustainedpain relief after final procedure (N ¼ 8).Table S6B. Pain characteristics of individual patients with nosustained pain relief after final procedure (N ¼ 8).Table S7A. Characteristics of patients ineligible for our protocolizedtreatment protocol (N ¼ 16).Table S7B. Pain characteristics of patients ineligible for ourprotocolized treatment protocol (N ¼ 16).Table S8. Characteristics of patients stratified according to achievedpain relief after the final treatment they received.Supplementary material is linked to the online version of the paper atwww.kidney-international.org.

REFERENCES1. Neumann HP, Jilg C, Bacher J, et al. Epidemiology of autosomal-

dominant polycystic kidney disease: an in-depth clinical study forsouth-western Germany. Nephrol Dial Transplant. 2013;28:1472–1487.

2. Grantham JJ. Clinical practice. Autosomal dominant polycystic kidneydisease. N Engl J Med. 2008;359:1477–1485.

3. Bajwa ZH, Sial KA, Malik AB, Steinman TI. Pain patterns in patients withpolycystic kidney disease. Kidney Int. 2004;66:1561–1569.

4. Miskulin DC, Abebe KZ, Chapman AB, et al. Health-related quality of lifein patients with autosomal dominant polycystic kidney disease and CKDstages 1-4: a cross-sectional study. Am J Kidney Dis. 2014;63:214–226.

5. Loukas M, Klaassen Z, Merbs W, et al. A review of the thoracic splanchnicnerves and celiac ganglia. Clin Anat. 2010;23:512–522.

6. Bajwa ZH, Gupta S, Warfield CA, Steinman TI. Pain management inpolycystic kidney disease. Kidney Int. 2001;60:1631–1644.

7. Hogan MC, Norby SM. Evaluation and management of pain in autosomaldominant polycystic kidney disease. Adv Chronic Kidney Dis. 2010;17:e1–e16.

8. Casteleijn NF, Visser FW, Drenth JP, et al. A stepwise approach foreffective management of chronic pain in autosomal-dominant polycystickidney disease. Nephrol Dial Transplant. 2014;29 Suppl 4:iv142–iv153.

9. Tellman MW, Bahler CD, Shumate AM, et al. Management of pain inADPKD and anatomy of renal innervation. J Urol. 2015;193:1470–1478.

10. Casteleijn NF, de Jager RL, Neeleman MP, et al. Chronic kidney pain inautosomal dominant polycystic kidney disease: a case report ofsuccessful treatment by catheter-based renal denervation. Am J KidneyDis. 2014;63:1019–1021.

11. Shetty SV, Roberts TJ, Schlaich MP. Percutaneous transluminal renaldenervation: a potential treatment option for polycystic kidney disease-related pain? Int J Cardiol. 2013;162:e58–e59.

12. Elzinga LW, Barry JM, Torres VE, et al. Cyst decompression surgery forautosomal dominant polycystic kidney disease. J Am Soc Nephrol. 1992;2:1219–1226.

13. Dunn MD, Portis AJ, Naughton C, et al. Laparoscopic cystmarsupialization in patients with autosomal dominant polycystic kidneydisease. J Urol. 2001;165:1888–1892.

14. Haseebuddin M, Tanagho YS, Millar M, et al. Long-term impact oflaparoscopic cyst decortication on renal function, hypertension and paincontrol in patients with autosomal dominant polycystic kidney disease.J Urol. 2012;188:1239–1244.

15. Lee DI, Andreoni CR, Rehman J, et al. Laparoscopic cyst decortication inautosomal dominant polycystic kidney disease: impact on pain,hypertension, and renal function. J Endourol. 2003;17:345–354.

16. Suwabe T, Ubara Y, Mise K, et al. Suitability of patients with autosomaldominant polycystic kidney disease for renal transcatheter arterialembolization. J Am Soc Nephrol. 2016;27:2177–2187.

17. Rana MV, Candido KD, Raja O, Knezevic NN. Celiac plexus block in themanagement of chronic abdominal pain. Curr Pain Headache Rep.2014;18:394.

18. Garcea G, Thomasset S, Berry DP, Tordoff S. Percutaneous splanchnicnerve radiofrequency ablation for chronic abdominal pain. ANZ J Surg.2005;75:640–644.

9

c l i n i ca l i nves t iga t i on NF Casteleijn et al.: Chronic pain management in ADPKD

19. Eisenberg E, Carr DB, Chalmers TC. Neurolytic celiac plexus block fortreatment of cancer pain: a meta-analysis. Anesth Analg. 1995;80:290–295.

20. Fadl Elmula FE, Jin Y, Yang WY, et al. Meta-analysis of randomizedcontrolled trials of renal denervation in treatment-resistanthypertension. Blood Press. 2015;24:263–274.

21. Chapuis O, Sockeel P, Pallas G, et al. Thoracoscopic renal denervation forintractable autosomal dominant polycystic kidney disease-related pain.Am J Kidney Dis. 2004;43:161–163.

22. Valente JF, Dreyer DR, Breda MA, Bennett WM. Laparoscopic renaldenervation for intractable ADPKD-related pain. Nephrol Dial Transplant.2001;16:160.

23. Gambaro G, Fulignati P, Spinelli A, et al. Percutaneous renal sympatheticnerve ablation for loin pain haematuria syndrome. Nephrol DialTransplant. 2013;28:2393–2395.

24. Casteleijn NF, Blais JD, Chapman AB, et al. Tolvaptan and kidney pain inpatients with autosomal dominant polycystic kidney disease: secondary

10

analysis from a randomized controlled trial. Am J Kidney Dis. 2016;63:1019–1027.

25. Mifflin KA, Kerr BJ. The transition from acute to chronic pain:understanding how different biological systems interact. Can J Anaesth.2014;61:112–122.

26. Bartley EJ, Fillingim RB. Sex differences in pain: a brief review of clinicaland experimental findings. Br J Anaesth. 2013;111:52–58.

27. McGuirk S, Fahy C, Costi D, Cyna AM. Use of invasive placebos in researchon local anaesthetic interventions. Anaesthesia. 2011;66:84–91.

28. Ware JE Jr. SF-36 health survey update. Spine (Phila Pa 1976). 2000;25:3130–3139.

29. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimateglomerular filtration rate. Ann Intern Med. 2009;150:604–612.

30. Sakakura K, Ladich E, Cheng Q, et al. Anatomic assessment ofsympathetic peri-arterial renal nerves in man. J Am Coll Cardiol. 2014;64:635–643.

Kidney International (2017) -, -–-