Calcium, Parathyroid Hormone, and Vitamin D: Major Determinants of Chronic Pain in Hemodialysis Patients

CJASN ePress. Published on July 2, 2009 as doi: 10.2215/CJN.00680109

Eliezer Golan,*† Isabelle Haggiag,* Pnina Os,* and Jacques Bernheim*† *Department of Nephrology and Hypertension, Meir Medical Center, Kfar Saba, and †Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel

Background and objectives: Pain is a frequent complaint of hemodialysis (HD) patients, yet information regarding its causes and frequency is relatively scarce. The aim of this study was to evaluate the frequency and possible causes of chronic pain in patients who are on long-term HD. Design, setting, participants, & measurements: We prospectively enrolled 100 patients who were undergoing maintenance HD for at least 3 mo. Pain was evaluated using the Brief Pain Inventory. Data collected on each participant included age, gender, ethnic origin, body mass index, smoking habits, time on dialysis, type of blood access, comorbidities, and biochemical and hematologic parameters. Results: The average age was 64.5 yr; the average time on dialysis 40.4 mo. Forty-five patients were male. Thirty-one participants were of Arabic origin. Fifty-three patients had diabetes, 36 of whom had diabetic retinopathy. Although 51 patients experienced chronic pain, only 19.6% described the pain as severe. Musculoskeletal pain, neuropathic pain, and headache were the most prevalent forms of pain. The presence of diabetic retinopathy and neuropathy (but not diabetes per se) and levels of intact parathyroid hormone, calcium, and calcitriol (but not 25-hydroxyvitamin D3) differed significantly between those who experienced chronic pain and those who did not. On a logistic regression model, higher serum calcium levels and intact parathyroid hormone levels >250 pg/ml were independently associated with chronic pain, as well as the presence of diabetic retinopathy. Calcitriol had a marginal effect. Conclusions: Disturbed mineral metabolism is strongly associated with chronic pain in long-term HD patients, along with microangiopathy. Clin J Am Soc Nephrol ●●: ●●●-●●●, 2009. doi: 10.2215/CJN.00680109

ain is a frequent complaint of hemodialysis (HD) pa- (DOPPS) by Bailie et al. (9) reported an undertreatment of pain tients (1–3), yet information regarding its origins, fre- in HD patients. In fact, pain was not treated adequately in the P quency, and management is relatively scarce. Most majority of patients. The purpose of this study was to evaluate published data come indirectly from studies focusing on the frequency and possible associations of chronic pain in pa- health-related quality of life (1,3). The reported frequency of tients who are on long-term HD. pain varies widely in these patients. Murtagh et al. (4), in a review of symptoms in ESRD, reported a weighted mean pain prevalence of 47%, with a range of 8 to 82%. Materials and Methods Although well-accepted guidelines are available for the man- We prospectively enrolled 100 patients who had been undergoing maintenance HD for at least 3 mo at the Meir Medical Center. The agement of cancer-related pain (5), no such recommendations institutional clinical research ethics review board approved the proto- exist for pain associated with HD. One review (6) suggested col. All patients provided written informed consent. using the same step-wise approach promulgated by the World The majority of patients were treated thrice weekly for 4 h. Cellulose- Health Organization to treat cancer pain; however, the treat- triacetate hollow fiber dialyzers (Sureflux150E or Sureflux 190E; Nipro ment of HD patients is complicated by the need to adjust Corp., Osaka, Japan) were used on AK200 or AK200S dialysis machines frequently the dosage of analgesic drugs and by increased risk (Gambro Ltd., Lund, Sweden) with blood flow of 280 to 350 ml/min for adverse effects (7,8). It is of no surprise, therefore, that an and dialysate flow of 500 ml/min. article from the Dialysis Outcomes and Practice Patterns Study At the time the study was performed, 67 patients had native arterio- venous fistula as their blood access, 24 had a polytetrafluoroethylene graft (GoreTex; W.L. Gore and Associates, Flagstaff, AZ), and nine had a tunneled central vein double-lumen catheter (PermCath; Quinton Received January 30, 2009. Accepted May 21, 2009. Instrument Co., Seattle, WA). Eighty-eight patients received intrave- Published online ahead of print. Publication date available at www.cjasn.org. nous recombinant human erythropoietin, either Epoetin Alfa (Eprex; E.G. and I.H. contributed equally to this work Cilag Ltd., Schaffhausen, Switzerland) or Epoetin Beta (Recormon; Hoffmann-La Roche Ltd., Basel, Switzerland), and intravenous iron Correspondence: Dr. Eliezer Golan, Department of Nephrology and Hyperten- therapy (Venofer; Vifor Int., St. Gallen, Switzerland). sion, Meir Medical Center, 59 Tchernichovsky Street, Kfar Saba 44281 Israel. Phone: ϩ972-9-747-2684; Fax: ϩ972-9-741-6918; E-mail: [email protected], The data collected on each participant included age, gender, origin, [email protected] body mass index, smoking habits, time on dialysis, type of blood access,

Copyright © 2009 by the American Society of Nephrology ISSN: 1555-9041/●●●●–0001 2 Clinical Journal of the American Society of Nephrology Clin J Am Soc Nephrol ●●: ●●●-●●●, 2009 comorbidities, and biochemical and hematologic parameters. Comor- classified as mild (1 to 4 points), moderate (5 to 6 points), or severe (7 bidities included malignancy, hypertension, ischemic heart disease to 10 points) (20). (IHD), stroke, peripheral arterial disease (PAD), diabetes, diabetic ret- Chronic pain was defined as pain of Ͼ3 mo duration (20). Pain inopathy, and neuropathy. The laboratory profile included serum cal- related to repeated dialysis access cannulation was assessed separately, cium, phosphorus, intact parathyroid hormone (iPTH), 25-hydroxyvi- including questions regarding the regular use of local anesthetic spray tamin D3 [25(OH)D3] 1,25-dihydroxyvitamin D3 (calcitriol), albumin, (ethyl chloride, various manufacturers) or cream (eutectic mixture of hemoglobin, iron, vitamin B12, folic acid, ferritin, glucose, and C-reac- lidocaine and prilocaine, EMLA 5%; AstraZeneca, London, UK). tive protein. For patients with diabetes, we also measured glycated hemoglobin (HbA ). The calcium-phosphorus product was calculated 1c Statistical Analysis for each patient, as well as Kt/V , using the second-generation urea Descriptive statistics including frequencies and means Ϯ SD were Daugirdas formula (10). Calcium levels were corrected for albumin. calculated for demographic variables. Log transformation was used for Levels of vitamin D [25(OH)D and calcitriol] were determined by 3 variables with a skewed distribution. ␹2, Fisher test, t tests, and logistic radioimmunoassay (DiaSorin, Stillwater, MN) and iPTH levels by im- regression using the forward stepwise method were used, as appropri- munoradiometric assay (N-TACT PTH SP IRMA; DiaSorin). A Hitachi ate. Statistical analysis was performed using SPSS 15 (SPSS, Chicago, 747 Clinical Analyzer was used to determine the serum levels of albu- IL). Statistical significance was set at 5% (P Ͻ 0.05). min, glucose, iron, calcium, and phosphorus. A Hitachi 917 was used for C-reactive protein, and Immulite 2000 (DCP, Los Angeles, CA) was used for ferritin. Access (Beckman, Fullerton, CA) was used for B12 and Results folic acid. Advia 120 (Bayer, Leverkusen, Germany) was used to deter- One hundred consecutive patients participated in the study. mine hemoglobin levels. Table 1 depicts the basic demographic characteristics and lab- All patients were interviewed by the same investigator (I.H.) during oratory values of the patients, and Table 2 shows the comor- an HD session. Pain was evaluated using the Brief Pain Inventory (BPI). bidities. This is an instrument for evaluating pain that assesses the intensity and Forty-five patients were male. There were no statistically characteristics of pain and determines the impact of pain on important aspects of a patient’s life. The BPI uses a 10-point scale whereby 0 ϭ “no significant differences in any of the parameters between male Ϯ pain” and 10 ϭ “pain as bad as you can imagine” to evaluate intensity and female patients except for Kt/V (male 1.27 0.18 female Ϯ Ͻ Ϯ of pain (11). The BPI has shown its validity across cultural and linguis- 1.45 0.27; P 0.001), 25(OH)D3 (male 22.0 10.8 female tic backgrounds (12–15) as well as diverse clinical situations (16–19), 16.1 Ϯ 8.7; P ϭ 0.004), and smoking habits (current/former: including dialysis patients (1,20). On the basis of the BPI scale, pain was male 4/8 female 4/1; P ϭ 0.018).

Table 1. Demographic characteristics and laboratory valuesa

Parameter Diabetes No Diabetes Total

n 53 47 100 Age (yr) mean Ϯ SD 64.8 Ϯ 10.1 64.1 Ϯ 17.2 64.5 Ϯ 13.9 range 43 to 90 26 to 87 26 to 90 Time on dialysis (mo) mean Ϯ SD 29.7 Ϯ 30.0 52.4 Ϯ 49.0b 40.4 Ϯ 42.0 range 3 to 168 6 to 204 3 to 204 Origin: Arab/non-Arab 22/31 9/38b 31/69 Smoker: current/former 4/8 4/1 8/9 BMI (kg/m2; mean Ϯ SD) 28.8 Ϯ 6.6 27.2 Ϯ 7.3 28.1 Ϯ 6.9 Kt/V (mean Ϯ SD) 1.34 Ϯ 0.23 1.40 Ϯ 0.26 1.36 Ϯ 0.23 Calcium (mg/dl; mean Ϯ SD) 9.3 Ϯ 0.5 9.6 Ϯ 0.6b 9.4 Ϯ 0.6 Phosphorus (mg/dl; mean Ϯ SD) 5.2 Ϯ 1.3 5.3 Ϯ 1.5 5.2 Ϯ 1.4 Calcium-phosphorus product (mg2/dl2; mean Ϯ SD) 48.1 Ϯ 11.8 50.8 Ϯ 15.1 49.3 Ϯ 13.5 iPTH (pg/ml; mean Ϯ SD) 109 Ϯ 93 187 Ϯ 180b 146 Ϯ 145 ␮ Ϯ Ϯ Ϯ b Ϯ 25(OH)D3 ( g/L; mean SD) 16.6 9.1 21.1 10.8 18.6 10.1 Calcitriol (pg/ml; mean Ϯ SD) 13.9 Ϯ 5.0 18.8 Ϯ 9.0b 16.1 Ϯ 7.4 Albumin (g/dl; mean Ϯ SD) 3.8 Ϯ 0.3 3.9 Ϯ 0.3 3.8 Ϯ 0.3 Hemoglobin (g/dl; mean Ϯ SD) 11.8 Ϯ 1.2 12.0 Ϯ 1.2 11.9 Ϯ 1.2 Iron (␮g/dl; mean Ϯ SD) 70.7 Ϯ 24.5 72.7 Ϯ 22.5 71.6 Ϯ 23.5 Ϯ Ϯ Ϯ Ϯ Vitamin B12 (ng/L; mean SD) 609.0 265.0 634.0 289.0 621.4 275.8 Ferritin (␮g/L; mean Ϯ SD) 1105.0 Ϯ 526.0 1180.0 Ϯ 689.0 1140.8 Ϯ 606.0 CRP (mg/dl; mean Ϯ SD) 2.04 Ϯ 3.56 1.46 Ϯ 1.47 1.77 Ϯ 2.80

a 25(OH)D3, 25-hydroxyvitamin D3; BMI, body mass index; CRP, C-reactive protein; iPTH, intact parathyroid hormone; b P Ͻ 0.05, diabetes versus no diabetes. Clin J Am Soc Nephrol ●●: ●●●-●●●, 2009 Chronic Pain in HD Patients 3

Table 2. Number of patients with comorbid conditionsa graphic and laboratory parameters between these two groups of patients with diabetes, although both groups had lower Parameter Male Female Total levels of calcium, iPTH, 25(OH)D3, and calcitriol compared n 45 55 100 with those without diabetes. Hypertension 39 48 87 Thirty-one patients were of Arabic origin. They had a signif- Diabetes 22 31 53 icantly higher rate of diabetes: 71% (22 of 31) compared with ϭ diabetic 14 22 36 44.9% (31 of 69) of non-Arabic patients (P 0.018). The per- retinopathy centage of HbA1c of patients who had diabetes and were of Ϯ Ϯ diabetic 10 18 28 Arabic origin was 7.6 1.6 compared with 6.6 1.3 in non- ϭ neuropathy Arabic patients with diabetes (P 0.016). Diabetic retinopathy IHD 21 25 46 was more prevalent in patients of Arabic origin: 54.8% (17 of 31) ϭ PAD 5 9 14 versus 27.5% (19 of 69) in non-Arabic patients (P 0.013). The CVA 7 11 18 difference in diabetic neuropathy approached but did not reach ϭ Malignancy 11 10 21 statistical significance (13 of 31 versus 15 of 69; P 0.054). colon cancer 6 2 8 Eighty-eight patients were receiving erythropoietin. The av- Ϯ breast cancer 0 3 3 erage dosage was 10,977 7758 U/wk. There was no statisti- other 5 5 10 cally significant difference between those who received erythropoietin and those who did not in the prevalence or intensity a There was no statistically significant difference in any of of chronic pain, as well as in any other demographic or labo- the comorbidities between male and female patients. CVA, ratory parameter. cerebrovascular accident; IHD, ischemic heart disease; PAD, peripheral arterial disease. A total of 51 patients experienced chronic pain. Table 3 depicts their demographic characteristics and laboratory values compared with the 49 patients who did not report chronic pain. Fifty-three patients had diabetes. Because patients with dia- Patients with chronic pain had statistically significant higher betes may have a different form of primary renal disease, levels of calcium and iPTH and lower levels of calcitriol [but diabetic nephropathy was considered to be the primary renal not 25(OH)D3] compared with those without chronic pain. disease only when it was biopsy proven or when diabetic More patients with chronic pain experienced diabetic retinop- retinopathy was also present and there was no evidence of athy and neuropathy compared with patients without chronic other renal disease. On the basis of this definition, the primary pain. None of the other comorbidities differed between the renal disease was considered to be diabetic nephropathy for 36 groups, including the incidence of diabetes. patients with diabetes. Twenty-five patients had nephrosclero- Of note, although none of the demographic parameters sis, nine had chronic glomerulonephritis, six had adult domi- reached statistical significance, patients with chronic pain nant polycystic kidney disease, four had chronic interstitial tended to be on HD for a longer period of time (48.0 Ϯ 50.2 nephritis, three had nephrolithiasis, and 17 had other or un- versus 32.5 Ϯ 29.7 mo; P ϭ 0.064). Smoking habits as well as the known type of renal disease. type of blood access did not differ between the two groups. Time on dialysis was significantly shorter for the 53 patients Diabetes was equally controlled among the patients with dia- with diabetes compared with those without diabetes: 29.7 Ϯ betes in the two groups, as indicated by the percentage of Ϯ Ͻ Ϯ 30.0 versus 52.4 49.0 mo (P 0.05). A higher percentage of HbA1c (7.1 1.5% in patients with diabetes and chronic pain patients with than without diabetes had cardiovascular comor- versus 7.0 Ϯ 1.5 in patients with diabetes and without chronic bidities: IHD 56.6 versus 34%, hypertension 96 versus 76.6%, pain). Some patients with chronic pain had much higher levels stroke 24.5 versus 10.6%, and PAD 22.6 versus 4%, respectively. of iPTH that were not found in the group without pain; this is The difference in IHD and PAD reached statistical significance reflected in the higher SD (180 versus 86) noted in Table 3. ϭ ϭ (P 0.04 and P 0.02, respectively). The average HbA1c was Ignoring the extreme values, however, had no significant im- 7.05 Ϯ 1.47% (7.20 Ϯ 1.50% for women and 6.80% Ϯ 1.40 for pact on the analysis of our findings. Table 4 depicts the odds men; the difference between genders was not statistically sig- ratio for chronic pain, calculated using a stepwise logistic re- nificant). Compared with those without diabetes, those with gression model. The independent variables found to be signif- diabetes had significantly lower levels of serum calcium, iPTH, icant were iPTH level Ͼ250 pg/ml, the presence of diabetic calcitriol, and 25(OH)D3. More patients with diabetes com- retinopathy, and a higher serum calcium level. Level of calcit- plained of chronic pain (56.6 versus 44.7%), but this difference riol Ͼ17 pg/ml demonstrated a marginal effect. was not statistically significant (Table 1). Thirty-six of the pa- Twenty-five (49%) patients described their pain as mild, 16 tients with diabetes had diabetic retinopathy. Compared with (31.4%) as moderate, and 10 (19.6%) as severe. The blood level the 17 with diabetes and without diabetic retinopathy, a signif- of calcium was the only parameter that demonstrated a corre- icantly higher percentage of them experienced chronic pain lation to the patients’ degree of pain (r ϭ 0.305, P ϭ 0.002). (72.2 versus 23.5%; P Ͻ 0.01), were younger (62.7 Ϯ 8.6 versus The 51 patients with chronic pain experienced it in diverse 69.4 Ϯ 11.5 yr; P ϭ 0.04), and were more obese (body mass organ systems: 28 (54.9%) reported chronic headache, 21 index 30.1 Ϯ 6.8 versus 25.9 Ϯ 5.1 kg/m2). There were no (41.2%) neuropathic pain, 13 (25.5%) back pain, 11 (21.6%) other statistically significant differences in any of the other demo- musculoskeletal pain, six (11.8%) chronic joint pain, and nine 4 Clinical Journal of the American Society of Nephrology Clin J Am Soc Nephrol ●●: ●●●-●●●, 2009

Table 3. Characteristics of patients with and without chronic pain

Chronic No Chronic Pa Characteristic Pain Pain

n 51 49 Gender: male/female 19/32 26/23 NS Age (yr; mean Ϯ SD) 64.0 Ϯ 12.5 65.0 Ϯ 15.2 NS Time on dialysis (mo; mean Ϯ SD) 48.0 Ϯ 50.2 32.5 Ϯ 29.7 NS Origin: Arab/non-Arab 17/34 14/35 NS Hypertension: yes/no 42/9 45/4 NS Diabetes: yes/no 30/21 23/26 NS Diabetic retinopathy: yes/no 26/25 10/39 0.002 Diabetic neuropathy: yes/no 23/28 5/44 0.001 IHD: yes/no 26/25 20/29 NS PAD: yes/no 9/42 5/44 NS CVA: yes/no 8/43 10/39 NS BMI (kg/m2; mean Ϯ SD) 28.9 Ϯ 7.1 27.2 Ϯ 6.7 NS Kt/V (mean Ϯ SD) 1.33 Ϯ 0.22 1.41 Ϯ 0.27 NS Malignancy: yes/no 11/40 10/39 NS Calcium (mg/dl; mean Ϯ SD) 9.6 Ϯ 0.7 9.3 Ϯ 0.5 0.008 Phosphorus (mg/dl; mean Ϯ SD) 5.3 Ϯ 1.3 5.1 Ϯ 1.4 NS Calcium-phosphorus product (mg2/dl2; mean Ϯ SD) 51.0 Ϯ 13.1 47.6 Ϯ 13.8 NS iPTH (pg/ml; mean Ϯ SD) 180 Ϯ 180 110 Ϯ 86 0.016 ␮ Ϯ Ϯ Ϯ 25 (OH)D3 ( g/L; mean SD) 17.4 9.4 19.9 10.6 NS Calcitriol (pg/ml; mean Ϯ SD) 14.3 Ϯ 6.0 18.0 Ϯ 8.0 0.01 Albumin (g/dl; mean Ϯ SD) 3.8 Ϯ 0.3 3.8 Ϯ 0.4 NS Hemoglobin (g/dl; mean Ϯ SD) 11.8 Ϯ 1.2 12.0 Ϯ 1.1 NS Iron (␮g/dl; mean Ϯ SD) 70.9 Ϯ 24.9 72.3 Ϯ 22.2 NS Vitamin B12 (ng/L; mean Ϯ SD) 653 Ϯ 283 588 Ϯ 267 NS Folic Acid (␮g/L; mean Ϯ SD) 8.7 Ϯ 2.4 9.2 Ϯ 1.9 NS Ferritin (␮g/L; mean Ϯ SD) 1172 Ϯ 650 1107 Ϯ 561 NS CRP (mg/dl; mean Ϯ SD) 1.52 Ϯ 1.41 2.02 Ϯ 3.71 NS

a NS, P Ͼ 0.05.

Table 4. OR for chronic paina needle insertion. There were no significant differences in demographic or laboratory parameters between patients who con- Parameter OR 95% CI sidered the needle insertion to be painful and those who did iPTH Ͼ250 pg/ml 9.99 1.88to53.08 not or between those who regularly used local anesthetics and Diabetic retinopathy 7.51 2.52to22.35 those who did not. Calcium mg/dl 3.70b 1.43to9.58 Calcitriol Ͼ17 pg/ 0.36 0.13to1.02 ml Discussion For Ͼ40 yr, dialysis has proved to be a successful life-sus- a CI, confidence interval; OR, odds ratio. taining therapy. As such, its effectiveness has been judged b When calcium changes in one unit (and the other mainly by patient survival; however, as this form of therapy independent variables remain the same), the log odds of matures and because the majority of patients are older and chronic pain will increase by 1.31. experience multiple comorbidities, health-related quality of life (HRQoL) becomes increasingly important. Pain is considered to (17.6%) abdominal pain. Seven (13.7%) patients had pain from be a highly relevant patient outcome in evaluating HRQoL in various other sources, such as phantom pain, steal syndrome, HD (1,8,21,22). Moreover, recent studies showed an association and nonspecific diffuse pain. between assessment of HRQoL and morbidity and mortality in Each needle insertion in patients with a fistula or graft as patients with ESRD (23). their blood access is expected to cause unavoidable pain; how- Our study showed that pain is common in patients who are ever, only 73 (80.2%) of 91 patients considered the procedure to on long-term HD; 51% of our patients experienced chronic pain. be painful at all, and only 37 (40.6%) believed that it was intense This is in agreement with other reports in the literature: Davi- enough to justify the regular use of a local anesthetic before son (1), in her publication on 205 Canadian HD patients, re- Clin J Am Soc Nephrol ●●: ●●●-●●●, 2009 Chronic Pain in HD Patients 5 ported that 50% of patients had chronic pain. In a review, within the recommended range and the cross-sectional nature Murtagh et al. (4) reported a weighted mean prevalence of pain of our study. of 47% but with a range from 8 to 82%. This wide range is Many of the previously published works on chronic pain in probably due to differences in the definition of chronic pain HD patients did not focus on the interrelation between pain and the method used to assess it, as well as differences in the and biochemical parameters, especially calcium homeostasis. In perception of pain among the diverse population studied. our study, calcium was an independent risk factor for chronic Although the prevalence of chronic pain in our study was pain, whereas calcitriol level had a marginal effect (Table 4). It almost identical to other reports, the severity of pain differed; should be noted that the difference in the average serum cal- for example, 55% of Davison’s patients rated their pain as cium levels between the groups with and without pain was severe (1), whereas only 19.6% of our patients did so. Differ- relatively small; however, there was strong statistical signifi- ences in the perception of pain, possibly based on cultural and cance (P ϭ 0.008; Table 3). Moreover, calcium level correlated ethnic factors, may explain these findings. Chiang et al. (24) with the severity of pain, and on a logistic regression model, described similar differences in HD patients in Taiwan as did higher serum calcium levels were independently associated Kimmel et al. (21) while discussing the impact of spiritual with chronic pain. Taken altogether, it seems that the relatively beliefs, psychosocial factors, and ethnicity on HRQoL and HD. small difference in the average serum calcium levels is of clin- Conversely, other than having a higher prevalence of diabetes ical significance. and retinopathy among patients of Arabic background, ethnic Of note, iPTH Ͼ250 pg/ml, well within the recommended origin by itself had no correlation with pain in our cohort. level for HD patients (150 to 300 pg/ml), is an independent risk Pain of musculoskeletal origin was the most frequent form of factor for chronic pain. Since the publication of the Kidney chronic pain reported by patients who were on maintenance Disease Outcomes Quality Initiative (K/DOQI) guidelines for HD (8). The findings in our study were of no exception, but it bone metabolism and disease in CKD (32), there has been an should be noted that other forms of pain were frequent as well, ongoing discussion regarding the optimal treatment of renal emphasizing the complex nature of chronic pain in HD pa- bone disease, including the optimal levels of PTH. Although tients. One study that focused on gastrointestinal symptoms in the focus is mainly on bone parameters as well as survival (33), dialysis patients found that abdominal pain was the most fre- our results add a different aspect that should be taken into quent gastrointestinal symptom, reported by 72% of HD pa- account, which is the possible adverse effect on chronic pain if tients (25). higher levels of PTH will be accepted. Headache is very common, usually restricted to the HD Noordzij et al. (34) analyzed data from the prospective Neth- session. We also found that a substantial number of patients erlands Cooperative Study on the Adequacy of Dialysis (NE- experience chronic headache, not triggered by the dialysis treat- COSAD) and reported that disturbed mineral metabolism was ment. This is in accordance with other reports. For example, the associated with muscle pain and cramps in dialysis patients. 2003 prospective study on headache in HD by Antoniazzi et al. Compared with our results, muscle pain was more prevalent in (26) reported that eight (28.6%) of 28 patients who did not their study, found in 68% of their patients, and the rate in- experience headache before dialysis developed this form of creased to 81% during 4 yr of follow-up. In agreement with our chronic headache while on HD. findings, there was a correlation between muscle pain and high Most published studies have found statistically significant calcium and iPTH levels. Noordzij et al. also described a corre- positive correlations between time on dialysis and chronic pain lation between phosphorus levels and pain; a correlation that (1). We observed the same trend in our data, but it did not reach we were unable to demonstrate. This difference may result statistical significance, probably because of the wide range of from the tight control of phosphorus in almost all of our pa- time (3 to 204 mo) that our patients were on long-term HD. tients, both with and without pain, as well as from the different Fifty-three of our patients had diabetes, which, by itself, had study methods, longitudinal versus cross-sectional, and differ- no impact on the presence of chronic pain; however, we found ent estimates of pain. a strong correlation between chronic pain and diabetic neurop- Because they have a pivotal role in bone metabolism, the athy and retinopathy. Microangiopathy is an established culprit correlation of calcium, PTH, and calcitriol with pain may seem in the complex pathogenesis of diabetic neuropathy (27). This obvious, because musculoskeletal pain is regarded as the most may point to a significant role of microangiopathy in the cre- prevalent form of chronic pain in HD patients; however, many ation and perpetuation of chronic pain in these patients. of our patients frequently experienced other forms of pain, so it Our patients with diabetes had significantly lower levels of seems that calcium, PTH, and calcitriol may be involved in calcium, PTH, 25(OH)D3, and calcitriol; this is in agreement chronic pain in these patients beyond their role in bone metab- with other reports (28,29). Low calcium and vitamin D are olism. Our work points out the likelihood that calcium, PTH, associated with endothelial dysfunction (30). We speculate that and calcitriol play a profound role in the chronic pain experi- the lower levels of calcium and calcitriol observed in our pa- enced by HD patients. Indeed, calcium (35) and vitamin D (36) tients may play a role in the development of chronic ischemic have been implicated as being directly involved in pain. pain in various organs by their impact on endothelial dysfunc- The cross-sectional nature of our study as well as the relation. In contrast to others (31), we could not demonstrate a tively small number of participants, although pointing to a correlation between hemoglobin level and chronic pain. prob- strong association of calcium metabolism with chronic pain in ably because the hemoglobin level of most of our patients was HD patients, precludes drawing firm cause-and-effect conclu- 6 Clinical Journal of the American Society of Nephrology Clin J Am Soc Nephrol ●●: ●●●-●●●, 2009 sions from our data. Nevertheless, we believe that our results pain assessment tool in Korea: A validation study of a provide a solid basis for studies that will further explore these Korean version of the Brief Pain Inventory. Oncology 66: relationships. The ultimate goal should be a better understand- 439–444, 2004 ing and treatment of chronic pain in HD patients. 15. Ger LP, Ho ST, Sun WZ, Wang MS, Cleeland CS: Valida- tion of the Brief Pain Inventory in a Taiwanese population. Acknowledgments J Pain Symptom Manage 15: 316–322, 1999 16. Poundja J, Fikretoglu D, Guay S, Brunet A: Validation of This work was presented as an abstract at the annual meeting of the the French version of the Brief Pain Inventory in Canadian American Society of Nephrology; October 31 through November 5, veterans suffering from traumatic stress. J Pain Symptom 2007, San Francisco, CA; and at the ERA-EDTA Congress; June 21 to 24, 2007; Barcelona, Spain. Manage 33: 720–726, 2007 17. Girard F, Chouinard P, Boudreault D, Poirier C, Richard C, Ruel M, Ferraro P: Prevalence and impact of pain on the Disclosures quality of life of lung transplant recipients. Chest 130: 1535– None. 1540, 2006 18. Lee MA, Walker RW, Hildreth TJ, Prentice WM: A survey References of pain in idiopathic Parkinson’s disease. J Pain Symptom Manage 32: 462–469, 2006 1. Davison SN: Pain in hemodialysis patients: Prevalence, 19. Ehde DM, Osborne TL, Hanley MA, Jensen MP, Kraft GH: cause, severity and management. Am J Kidney Dis 42: 1239– The scope and nature of pain in persons with multiple 1247, 2003 sclerosis. Mult Scler 12: 629–638, 2006 2. Mercadante S, Ferrantelli A, Tortorici C, Lo Cascio A, Lo 20. Davison SN, Jhangri GS: The impact of chronic pain on Cicero M, Cutaia I, Parrino I, Casuccio A: Incidence of chronic pain in patients with end-stage renal disease on depression, sleep, and the desire to withdraw from dialysis dialysis. J Pain Symptom Manage 30: 302–304, 2005 in hemodialysis patients. J Pain Symptom Manage 30: 465– 3. Shayamsunder AK, Patel SS, Jain V, Peterson RA, Kimmel 473, 2005 PL: Sleepiness, sleeplessness, and pain in end-stage renal 21. Kimmel PL, Emont SL, Newmann JM, Danko H, Moss AH: disease: Distressing symptoms for patients. Semin Dial 18: ESRD patient quality of life: Symptoms, Spiritual beliefs, 109–118, 2005 psychosocial factors, and ethnicity. Am J Kidney Dis 42: 4. Murtagh FE, Addington-Hall J, Higginson IJ: The preva- 713–721, 2003 lence of symptoms in end stage renal disease: A systematic 22. Wu AW, Fink NE, Marsh-Manzi JV, Meyer KB, Finkelstein review. Adv Chronic Kidney Dis 14: 82–99, 2007 FO, Chapman MM, Powe NR: Changes in quality of life 5. Zech DF, Grond S, Lynch J, Hertel D, Lehmann KA: Vali- during hemodialysis and peritoneal dialysis treatment: Ge- dation of World Health Organization guidelines for cancer neric and disease specific measures. J Am Soc Nephrol 15: pain relief: A 10-year prospective study. Pain 63: 65–76, 743–753, 2004 1995 23. Kimmel PL, Patel SS: Quality of life in patients with 6. Ahmad S, Barakzoy AS, Moss AH: Efficacy of the World chronic kidney disease: Focus on end-stage renal disease Health Organization analgesic ladder to treat pain in end- treated with Hemodialysis. Semin Nephrol 26: 68–79, 2006 stage renal disease. J Am Soc Nephrol 17: 3198–3203, 2006 24. Chiang CK, Peng YS, Chiang SS, Yang CS, He YH, Hung 7. Cohen LM, Moss AH, Weisbord SD, Germain MJ: Renal KY, Wu KD, Wu MS, Fang CC, Tsai TJ, Chen WY: Health- palliative care. J Palliat Med 9: 977–992, 2006 related quality of life of hemodialysis patients in Taiwan: A 8. Davison SN: Chronic pain in end-stage renal disease. Adv multicenter study. Blood Purif 22: 490–498, 2004 Chronic Kidney Dis 12: 326–334, 2005 25. Cano AE, Neil AK, Kang JY, Barnabas A, Eastwood JB, 9. Bailie GR, Mason NA, Bragg-Gresham Jl, Gillespie BW, Nelson SR, Hartley I, Maxwell D: Gastrointestinal symp- Young EW: Analgesic prescription patterns among hemo- toms in patients with end-stage renal disease undergoing dialysis patients in the DOPPS: Potential for underpre- treatment by hemodialysis or peritoneal dialysis. Am J scription. Kidney Int 56: 2419–2425, 2004 Gastroenterol 102: 1990–1997, 2007 10. Daugirdas JT: Second generation logarithmic estimates of 26. Antoniazzi AL, Bigal ME, Bordini CA, Tepper SJ, Speciali single-pool variable volume Kt/V: An analysis of error. JG: Headache and hemodialysis: A prospective study. J Am Soc Nephrol 4; 1205–1213, 1993 Headache 43: 99–102, 2003 11. Larue F, Colleau SM, Brasseur L, Cleeland CS: Multicentre study of cancer pain and its treatment in France. BMJ 310: 27. Zochodne DW: Diabetes mellitus and the peripheral ner- 1034–1037, 1995 vous system: Manifestations and mechanisms. Muscle 12. Klepstad P, Loge JH, Borchgrevink PC, Mendoza TR, Clee- Nerve 36: 144–166, 2007 land CS, Kaasa S: The Norwegian Brief Pain Inventory 28. Martinez I, Saracho R, Moina I, Montenegro J, Llach F: Is questionnaire: Translation and validation in cancer pain there a lesser hyperparathyroidism in diabetic patients patients. J Pain Symptom Manage 24: 517–525, 2002 with chronic renal failure? Nephrol Dial Transplant 13[Suppl 13. Mystakidou K, Mendoza T, Tsilika E, Befon S, Parpa E, 3]: 9–11, 1998 Bellos G, Vlahos L, Cleeland C: Greek Brief Pain Inventory: 29. Pittas AG, Lau J, Hu FB, Dawson-Hughes B: The role of Validation and utility in cancer pain. Oncology 60: 35–42, vitamin D and calcium in type 2 diabetes: A systematic 2001 review and meta-analysis. J Clin Endocrinol Metab 92: 2017– 14. Yun YH, Mendoza TR, Heo DS, Yoo T, Heo BY, Park HA, 2029, 2007 Shin HC, Wang XS, Cleeland CS: Development of a cancer 30. Wang TJ, Pencina MJ, Booth SL, Jacques PF, Ingelsson E, Clin J Am Soc Nephrol ●●: ●●●-●●●, 2009 Chronic Pain in HD Patients 7

Lanier K, Benjamin EJ, D’Agostino RB, Wolf M, Vasan RS: mended intact PTH levels do not prevent low-turnover Vitamin D deficiency and risk of cardiovascular disease. bone disease in hemodialysis patients. Kidney Int 73: 771– Circulation 117: 503–511, 2008 777, 2008 31. Plantinga LC, Fink NE, Jaar BG, Huang IC, Wu AW, Meyer 34. Noordzij M, Boeschoten EW, Bos WJ, Dekker FW, Bossuyt KB, Powe NR: Relation between level or change of hemo- PM, Krediet RT, Korevaar JC, NECOSAD Study Group: globin and generic and disease-specific quality of life mea- Disturbed mineral metabolism is associated with muscle sures in hemodialysis. Qual Life Res 16: 755–765, 2007 and skin complaints in a prospective cohort of dialysis 32. National Kidney Foundation: K/DOQI clinical practice patients. Nephrol Dial Transplant 22: 2944–2949, 2007 guidelines for bone metabolism and disease in chronic 35. Gribkoff VK: The role of voltage-gated calcium channels in kidney disease. Am J Kidney Dis 42[Suppl 3]:S1–S201, 2003 pain and nociception. Semin Cell Dev Biol 17: 555–564, 2006 33. Barreto FC, Barreto DV, Moyse´s RM, Neves KR, Canziani 36. Mascarenhas R, Mobarhan S: Hypovitaminosis D-induced ME, Draibe SA, Jorgetti V, Carvalho AB: K/DOQI-recom- pain. Nutr Rev 62: 354–359, 2004