Abstract
Background
Several studies have reported dialysis-related headache in adults. We investigated headache and its characteristics in pediatric and adolescent patients with chronic kidney disease and patients treated with dialysis, and compared characteristics of patients with and without headache in the entire cohort and separately among dialysis and among chronic kidney disease patients.
Methods
Patients and their parents who attended a nephrology clinic and hemodialysis unit were interviewed regarding the existence of headache and its characteristics. We reviewed hospital files for medical history, blood test results, and pharmacologic treatment. Headache was defined according to International Headache Society criteria.
Results
The cohort comprised 60 patients: 39 with chronic kidney disease without hemodialysis and 21 treated with hemodialysis; 39 were males, mean age 11.9 ± 5.3 years. Twenty-six (43.3%) reported experiencing headaches. The hemodialysis group had a higher rate of headache than the chronic kidney disease patients, at 76.2% vs. 25.5%, p < 0.001.
In the hemodialysis group, 15 out of 16 reported dialysis-related headache; 14 (87.5%) of these had migraine characteristics. For the entire cohort, headache was associated with hemodialysis, chronic kidney disease grade, lower glomerular filtration rate anemia and a higher parathyroid hormone level. In logistic regression analysis, glomerular filtration rate was significantly associated with headache, odds ratio 2.74 (95% CI 1.56–4.82, p < 0.001).
Conclusions
A high rate of headache, mostly migraine type, was reported by hemodialysis patients. Hemodialysis, anemia, higher parathyroid hormone levels, phosphate, and lower glomerular filtration rate are strongly associated with headache among chronic kidney disease pediatric and adolescent patients.
Introduction
The prevalence of chronic kidney disease has increased fourfold in the last 30 years (1). Chronic kidney disease affects virtually every organ system and thus has a major impact not only on mortality but also on quality of life (1). Dialysis-related headache was first described by Bana et al. in 1972 as resulting from changes in the balances of fluid and electrolytes during dialysis (2). However, more recent studies suggest other mechanisms, none of which has been elucidated. Variations in urea, sodium, magnesium, and hemodynamic parameters, as well as blood pressure levels and body weight changes, are among the factors proposed to be associated with headache in this setting (3,4).
Studies that have been published regarding dialysis-related headache have been performed only in adult populations (1–9). We did not identify any published studies of headaches among pediatric or adolescent patients with chronic kidney disease or on hemodialysis. The aims of this study are firstly to define the existence, incidence and character of headaches in pediatric patients with chronic kidney disease, and to compare differences in electrolyte balance, the proportion of dialysis patients, and blood pressure values between patients with and without headache. Secondly, the study aims to evaluate, among hemodialysis patients and among chronic kidney disease patients, headache character and incidence, and possible differences in electrolyte balance between those with and without headaches.
Methods
The study group comprised children, adolescents and young adults aged 3 to 21 years with chronic kidney disease or with end-stage chronic kidney disease maintained on dialysis for at least two months during 2015–2016, in the dialysis unit of the Nephrology Institute of Schneider Children’s Medical Center of Israel, which is a tertiary, university-affiliated, pediatric medical center. All patients who arrived at the unit with chronic kidney disease during the study period were enrolled in the study.
The diagnosis of headache type was based on International Headache Society ICHD-3, beta version criteria (10). The study was approved by the Research Ethics Board of Rabin Medical Center (approval no. RMC 0511-14). Parents of patients under the age of 18 and patients older than 18 gave their informed consent before participation in the study.
Headache diagnosis
The study participants were interviewed in the Pediatric Nephrology Institute by the head of the pediatric headache clinic in our hospital. Children/young adults and their parent(s), or only the parent(s) of very young children, were interviewed regarding the number of headaches per month, the nature of the headache attacks, the duration of the headaches and related symptoms such as vomiting, photophobia and phonophobia. Dialysis patients and their parents were interviewed regarding their tolerance of hemodialysis treatment and its possible relation to headaches. Dialysis headache and other types of headache were diagnosed according to the ICHD-3 beta version criteria (10). For very young children with limited verbal communication, headache symptoms were determined by the children’s complaints and parents’ impressions of children’s behavior (according to the criteria of the ICHD-3 beta version) (10). The medical files were reviewed to access information regarding chronic kidney disease and its duration, the background of other diseases, medications taken, and blood tests. In addition, parents were interviewed about their personal medical history of migraine (yes/no; diagnosed/not diagnosed by a physician) and characteristics of headaches they experienced.
Laboratory investigations
Blood samples are routinely taken during visits to the nephrology department for patients with chronic kidney disease, or once monthly in the hemodialysis unit before beginning hemodialysis. The following blood tests were evaluated in our study: Blood count, bicarbonate, sodium, potassium, urea alkaline phosphatase, magnesium, calcium, phosphate, iron, iron saturation and parathyroid hormone. Total iron/transferring saturation was calculated as follows: Iron (mcq/dL)/ transferrin (mg/dL) × 70.9% (<20% indicates iron deficiency) (11).
Dialysis procedure
All the study participants were being treated by hemodialysis three times per week, except for two patients with residual renal function who were being treated two times per week. The hemodialysis procedure was performed using only bicarbonate dialysate.
Definitions
Chronic kidney disease grades 1–5 were defined according to estimated glomerular filtration rate (12). Glomerular filtration rate was calculated according to the Schwartz formula for patients up to 17 years (13) and the GFR MDRD equation (Modification of Diet in Renal Disease study) for patients older than 17 years (14). Ultrafiltration volume was defined as the amount of fluid measured in liters that is removed from blood during the dialysis procedure. The ultrafiltration rate/predialysis weight (%) was defined as the ultrafiltration volume per hour divided by a patient’s weight before dialysis in kilograms (%). Abnormal/high blood pressure was diagnosed according to accepted definitions (15,16).
Statistical analysis
We performed three analyses of the data as follows: All patients with all grades of chronic kidney disease including dialysis patients with and without headaches; chronic kidney disease patients only, and hemodialysis patients only. The data were analyzed using BMPD software (17). A sample size calculation prior to conducting the study showed that 40 patients with chronic kidney disease and 20 patients on hemodialysis would be required to yield a power of 97.5% for a statistically significant result. This computation assumed that the difference in headache proportions is −0.50 (specifically, 0.25 vs. 0.75).
Continuous variables were calculated as means and standard deviations. Since the sample size was relatively small, and since some parameters did not have a Gaussian distribution, the nonparametric Mann Whitney U-test was used to compare between groups. Discrete variables were calculated as numbers and percentages, and compared between groups using Pearson’s chi-square or Fisher’s exact test, as applicable. A p value of ≤ 0.05 was considered significant. Stepwise logistic regression was used to identify the variables most significantly associated with headache-associated parameters.
Results
Characteristics of the study group
Sixty patients (39 males and 21 girls), mean age 11.9 ± 5.3 years, range 3–21 years, median 13 years, were included in the study. Four patients were over 18 years old. Thirty-nine patients had chronic kidney disease, and 21 had end-stage chronic kidney disease and were on chronic hemodialysis. The estimated glomerular filtration rate was in the range of 4–85 mL/min/1.73 m2, median 25.5 mL/min/1.73 m2. Thirty-one patients (51.7%) had anemia. Iron deficiency was found in four out of 60 (6.7%) of all the patients. Headache was reported by 26 (43.3%) of the patients in the cohort.
Clinical and demographic parameters of 39 chronic kidney disease patients and 21 hemodialysis treated patients.
Values are presented as n (%) or as mean ± SD; mos: months.
Comparing patients with and without headaches
Clinical parameters and blood tests of patients without and with headaches.
Values are presented as n (%) or as mean ± SD.
In a logistic regression, GFR was associated with headache, odds ratio 2.74 (95% CI 1.56–4.82, p < 0.001). The area under the curve was 0.79.
Hemodialysis patients
Characteristics of pediatric patients on hemodialysis, according to the presence of headaches.
Values are presented as n (%) or as mean ± SD.
Of the 16 (76.2%) dialysis patients who reported headaches, 14 (87.5%) had migraine-type headaches; two (12.5%) had tension-type headaches. Five patients reported predialysis headaches and reported subsequently experiencing headaches only during dialysis sessions, and at an increasing frequency. These headaches were defined in two ways: As primary headaches (migraine or tension), which describes the headache type before dialysis, and as dialysis headaches according to the ICHD-3 criteria (10). One of the five patients with a pre-dialysis headache did not have headaches during the dialysis sessions.
The 23.8% (five out of 21) rate of headaches before the dialysis procedure was similar to the rate for chronic kidney disease patients who were not treated with hemodialysis (10 out of 39 [25.6%], p = 0.98). In the hemodialysis group, six patients had migraine-type headaches without aura, eight had migraine-type headaches with aura, and two had tension-type headaches. Of the five patients who reported pre-dialysis headaches, four had migraines and one had tension headaches. All five patients reported migraine-type headaches after beginning treatment with hemodialysis. Three patients had undergone kidney transplants, with subsequent rejections, and then returned to hemodialysis treatment. These patients reported that their headaches disappeared during the period when their kidney transplants were functioning, and that the headaches returned with hemodialysis.
The locations of the headaches were: Frontal, 6; frontal temporal, 1; temporal, 6; parietal, 1. Two patients could not define the location of their headaches. Regarding headache character, six patients had pressing pain, eight pulsating/throbbing pain and two could not describe their pain. Eleven patients (68.7%) reported both phonophobia and photophobia.
Unilateral pain headache was reported by four patients, bilateral pain by six, and six patients could not respond regarding the side of the headache. Eleven (68.7%) of the16 hemodialysis patients with headaches reported that their headaches began only during the second or third hour of a dialysis session. One patient (6.3%) reported headaches starting towards the end of the session, and the rest reported variations in the timing of headache initiation.
Chronic kidney disease patients
Ten of 39 (25.6%) patients with chronic kidney disease reported headaches: six reported migraine (three with aura and three without aura), and four reported tension-type headaches. Anemia and a higher phosphate blood level were the only parameters associated with headaches in this sub-group of patients. The proportions of anemia in patients with headaches vs. patients without headaches were 70.0% and 27.6%, respectively (p = 0.047). Mean phosphate levels were 5.3 ± 0.7 mg/dL and 4.6 ± 0.8 mg/dL, respectively, p = 0.02.
Missing data and exclusion of patients from the analysis
All patients and/or their parents who attended the chronic kidney disease clinic or dialysis unit agreed to participate in the study and were interviewed. According to the regulations of our hospital Helsinki committee, one patient could not participate since she was taking part in another study. Two patients who were older than 21 years were also excluded. Only a few parameters were missing for patients who participated in the study. Among chronic kidney disease patients without dialysis, data were missing for phonophobia (one patient), and unilateral pain (one patient), bicarbonate tests were missing for two and PTH for one. Among chronic kidney disease, parental migraine information was missing in two patients.
Discussion
The main finding of this study is the higher frequency, by three times, of headaches among hemodialysis pediatric and adolescent patients than among patients with end stage chronic kidney disease.
1. Patients with and without headaches
Considering all the patients in the cohort, 16 (61.5%) of the 26 with headaches were treated with hemodialysis. Headaches were significantly associated with higher values of the following parameters: Hemodialysis rate; chronic kidney disease grade; lower GFR, blood levels of urea, ferritin, parathyroid hormone and phosphate; the number of concomitant medications for chronic kidney disease taken per day; and the proportion with anemia (Table 2). The mean number of medications was higher for patients on hemodialysis; however, the headaches do not seem to be related to the side effects of the medications, since no statistically-significant difference was found between those with and without headaches for the mean number of medications taken that may cause headaches.
Anemia was found more often in the patients reporting headaches, concurring with other studies that showed associations between anemia and headaches in children and adolescents (18,19). All four of our patients with iron deficiency anemia reported headaches. Compared to patients without headaches, patients with headaches had lower glomerular filtration rates and higher chronic kidney disease grade, both of which have been associated with a higher probability of electrolyte imbalance and anemia (20). During dialysis, erythrocyte and iron homeostasis physiology change and levels of ferritin and hepcidin, a peptide produced by the liver, increase (20,21). Elevated hepcidin is increased in inflammation and decreased in iron reabsorption (21). A higher mean ferritin level among patients with headaches may indicate an inflammation process that is caused by chronic kidney disease (20,21). Proinflammatory factors such as interleukin (IL)-1, soluble tumor necrosis factor receptor 1 and IL- 6 were found to be elevated in pediatric migraineurs and in pediatric hemodialysis patients (22,20,21).
In the current study, abnormal blood pressure was not associated with headache in chronic renal failures patients. This contrasts with Göksan et al.’s study (3), which reported an association between elevated blood pressure and predialysis headaches. Headache is only caused by hypertension if there is a hypertensive crisis (10). According to the ICHD-3 criteria, headache can be caused by arterial hypertension, usually during an acute rise in systolic and/or diastolic (blood pressure above the 99th percentile); it remits after normalization of blood pressure. Since our patients are frequently monitored and receive pharmacologic treatments, we assume that extreme high blood pressure is not the reason for their headaches. Our finding of a higher level of parathyroid hormone among pediatric patients with headaches supports the report of higher levels of this hormone in hemodialysis patients with chronic pain that included headaches, albeit in an adult population (23). Since many hemodialysis patients frequently experience musculoskeletal and other forms of pain such as headaches, the authors of that study concluded that calcium, parathyroid hormone and calcitriol may be involved in chronic pain in this setting, beyond their role in bone metabolism. We note that although significantly higher among patients with than without headaches, the mean level of phosphate was in the normal range according to our laboratory (3.3–5.4 mg/dL), even for the patients with headaches.
2. Hemodialysis and headache
The current study showed an association of hemodialysis with headache. However, we did not find higher levels of predialysis sodium, potassium or urea levels among adult patients with dialysis headache, in contrast to the findings of another study (3). The lack of association between dialysis headache and phosphate levels reported herein concurs with the findings of another study (23). Although the blood magnesium level tended to be lower in adult dialysis patients with headaches than in those without headaches (4), the magnesium blood level was within the normal range in both groups, and we presume that this was not the cause of headache. In our pediatric hemodialysis unit, all patients were treated with bicarbonate dialysis. Neyer et al. (24) reported more frequent and severe headaches during the procedure among adult patients who received acetate dialysis than in those who received bicarbonate.
The headache in our dialysis patients may result from a mild form of dialysis disequilibrium syndrome. This syndrome is characterized by neurological symptoms of headache, nausea, vomiting, muscle cramps and tremor, due to the rapid removal of urea during hemodialysis. Dialysis disequilibrium syndrome develops primarily from an osmotic gradient of urea that forms between the brain and the plasma as a result of rapid hemodialysis (25–27). The difference in urea between dialysis patients with and without headaches was similar, though the numbers of patients were small. Intravascular volume depletion during dialysis has been reported to result also from an imbalance between the rates of extracorporeal ultrafiltration and refilling of the blood compartment from the interstitial space. Only two of our hemodialysis patients needed fluids routinely during their hemodialysis sessions, due to low symptomatic blood pressure (27). None of the patients in this cohort reported the full clinical symptoms of disequilibrium syndrome.
All except one of our hemodialysis patients with headaches related their headaches to dialysis; this includes the five patients with predialysis headaches. Thus, we assume that the procedure of dialysis itself triggers headaches during the sessions. Patients with headaches had a higher ultrafiltration rate (l/hour)/predialysis weight/kg (%): median values were 9.0 vs. 6.7%. Jain et al. (28) found that hypotension, headache and cramping reported in pediatric hemodialysis patients occurred when blood volume changed relative to body weight by more than 8% in the first hour and by more than 4% during the rest of the session. Most (75%) of the patients in the current study who reported having headaches in the middle of dialysis (after 2–3 hours) had a higher ultrafiltration rate/pre-dialysis weight %, as was also reported in adults [5,8,9]. In the current study, 75% of the patients with hemodialysis headaches were males, compared with 65% of males in the entire cohort. This contrasts with the 40% (12/30 patients) males with headaches among the dialysis patients in Goskan et al's study [3; their cohort comprised 52.3% males.
Among our dialysis patients, eight (50%) had migraine-type headaches with aura, six (37.5%) had migraine-type headaches without aura and two (12.5%) had tension headaches. Similarly, among adults, the most frequently observed pattern of dialysis-related headache is the migraine pattern (8). The trigeminovascular system is activated in migraine attacks. Alessandri et al. (29) reported higher basal plasma concentrations of calcitonin gene-related peptide in patients (that is reported to be related to migraine pathophysiology (30,31), with dialysis-related headaches than in hemodialysis patients without headaches. They suggested that changes in the plasma concentrations of this peptide could contribute to the susceptibility of hemodialysis-related headache (32).
Patients with chronic kidney disease and headache
Among patients with chronic kidney disease, the mean level of blood phosphate was higher, and the proportion with anemia was greater among those with headaches than among those without headaches. Phosphate was not found to be one of the factors associated with chronic pain and headache in 100 adult hemodialysis patients (23). Although our chronic kidney disease patients had higher levels of phosphate, this level was still within the norms of our laboratory. The role of anemia in headaches is discussed above.
Study limitations
The main limitation of this study is the relatively small size of the cohort. However, pediatric chronic kidney disease is an infrequent condition (1), and this apparently explains the absence of studies on headaches in this population. Despite the relatively small size of the dialysis group, the 21 pediatric patients compromise 46% of the hemodialysis patients in our country. Psychological evaluations were not in the scope of the current study, despite our belief that psychological status may influence headaches in children with chronic disease. Adult patients treated with hemodialysis have been shown to have a higher probability of negative emotions, anxiety and mood disorder, and their subjective chronic complaints have been shown to be associated with psychological distress (32).
Conclusions
Headache is a frequent complaint in adolescent and pediatric hemodialysis patients, and is frequently associated with the dialysis procedure. Migraine-type headache is the most frequent type of headache in end stage kidney disease pediatric and adolescent patients treated by hemodialysis. Among pediatric and adolescent patients, headache was found to be more frequent in hemodialysis patients than in chronic kidney disease patients, and was strongly associated with anemia in all grades of chronic kidney disease, as well as higher parathyroid hormone levels, phosphate, and lower glomerular filtration rate. A prospective multicenter study is needed to verify and generalize our results, and to define the character and risk factors of dialysis and chronic kidney disease headache in children and adolescents.
Article highlights
Headaches are significantly more frequent in dialysis patients than in chronic kidney disease patients on dialysis. Most patients with dialysis headache have migraine character headaches. Headache is a frequent complaint among adolescent and pediatric hemodialysis patients. Among pediatric and adolescent patients, headache is associated with anemia in all grades of chronic kidney disease and low GFR.
Footnotes
Acknowledgement
The authors thank Ms Pnina Lilos for her statistical analysis and Ms Cindy Cohen for English language editing and the secretaries of the nephrology institute, Mrs Yana Spedt-Aizeman and Ms Lilia Korn-Berenshtein, for their assistance.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.