In this observational study, 17.6% of patients on hemodialysis therapy in Italy from 2015 to 2017 were potentially affected by CKD-aP. This prevalence is hard to compare with literature, which originates from qualitative studies using different ways of measuring CKD-aP and produced inconsistent data 2, 16. Nevertheless, the DOPPS phase 5 (2012-2015) 2 found from 5 to 20% of hemodialysis patients at least moderately bothered by pruritus. Facing the absence of a specific healthcare service identifying the CKD-aP and of clinical information (e.g. coming from dialysis registries, general practitioner’s database, laboratory values), we identified patients by means of the reimbursed supply of gabapentin, pregabalin, thalidomide, antihistamines, recommended by the current guidelines 6, 7, and performance of the UV phototherapy. The administrative healthcare databases record only the healthcare reimbursed by the INHS. Nevertheless, since none of them is specific for CKD-aP, we excluded some concomitant conditions (see methods) that frequently cause pruritus and can be treated with the aforementioned therapies 6, 7. Particularly, antihistamines corresponded to the most common therapeutic strategy for CKD-aP in the Italian clinical practice 2. Their chronic supply is reimbursed by the INHS in presence of a chronic severe condition (e.g. the excluded diseases), but also in case of seasonal allergic rhino-conjunctivitis that needs long-term treatment with antihistamines 17, which is, unfortunately, hard to identify within the administrative databases. Therefore, inevitably the few patients affected by this particular condition and coincidently by CKD felt into our CKD-aP cohort. Actually, we selected the CKD-aP patients within the limits of Italian administrative healthcare databases. CKD-aP subjects identified by this study were mostly males and elderly (mean age 69±13), in line with literature 1, 2, 8.
To date, very few researches have examined the prevalence, characteristics and outcomes of CKD-aP 16. Some of them have raised the critical frequent tendency of underreporting pruritus by patients and overlooking it by general practitioners, nephrologists and other healthcare professionals 1, 2, 8. The lack of renal functioning information and of possible previous lines of treatment prevents us from defining the severity of CKD. However, based on a study evaluating the reasons for underreporting pruritus in CKD patients 8, we can suppose that most of them received healthcare by the INHS in a very discomfort condition, probably affected by at least a moderate form of CKD-aP.
The etiopathogenesis of CKD-aP is still unknown and probably multifactorial, but some concomitant metabolic disorders (e.g. related to serum calcium, phosphorus and ferritin, parathormone, hemoglobin and albumin) have been frequently highlighted, though causality still has not been confirmed 1-4. In this study, the prevalence of hyperphosphatemia, hyperparathyroidism and anemia resulted higher in the CKD-aP cohort than in the non-CKD-aP one. Higher percentages of patients affected by the other comorbidities of interest were also found in the CKD-aP cohort. It is acknowledged that subjects suffering from CKD-aP are characterized by low physical and mental status, but it is still unclear among the concomitant diseases which ones are involved in the etiopathogenesis, or carried by patients from the pre-dialysis period or only consequences, since no supporting evidence still have been provided.
The CKD-aP symptom burden is high and it is very important to readily reduce it. Since an established and effective cure does not yet exist, clinicians’ attempt is based on the available evidence, even if not proven. The most updated guidelines 6, 7 do not recommend a clear management, but recent studies and reviews 2, 16, 18, 19 showed a tendency to follow a stepwise approach. It consists ideally in first trying to reach dialysis clearance target or mineral and bone ones; then to treat skin xerosis, if present, with emollient creams; if pruritus persists, to prescribe gabapentin or pregabalin. The dialytic approach will be discussed in the cost analysis section. Therapies for mineral and bone disorders are not evaluable through administrative databases, as well as the topical treatments 7.
In real life, nephrologists actually prescribe first and second generation antihistamines to treat CKD-aP, without further intervention. More than half of the medical directors interviewed by Rayner and colleagues prescribed oral and topical antihistamines as first choice for pruritus 2. In our study, antihistamines were dispensed to about 50% of CKD-aP patients, both one year before and after the index date. Since diagnoses of dermatitis or autoimmune pathologies possibly related to the prescription of antihistamines were excluded, we can state that the antihistamines supply were the greater marker of Italian CKD-aP patients in the ReS database. Moreover, in the same study 2, less than 10% of the interviewed patients affected by pruritus were treated with gabapentin or pregabalin. Particularly, even if gabapentin is, by now, the only drug worldwide marketed with the highest evidence against CKD-aP 20, in Italy it is not used as first choice 2. We found that gabapentin was supplied to 10% of CKD-aP patients. It is worth mentioning that its reimbursement is limited to specific cases of neuropathy 21, without ever mentioning the CKD-aP, and to epilepsy out of this recommendation. Therefore, the portion of subjects treated with gabapentin in this study resulted underestimated. This analysis also assessed the use of pregabalin, but no hemodialysis patient received it charged to the INHS. Thalidomide is also recommended by current guidelines 7, with a number of proposed mechanisms of action against pruritus, but it has not been reimbursed to any of our hemodialysis patients. UV phototherapy has shown dramatic improvement in pruritus 16 and in this study it was performed, charged to the INHS, to 1.4% of CKD-aP patients. The doubling of frequency from 2 to 4 UV therapies after the index date could suggest that it was successful. Based on the most updated literature, 4 theories on the pathophysiologic mechanisms behind CKD-aP exist 16: peripheral neuropathy, immune system dysregulation, opioid imbalance, and toxin deposition. The identification of comorbidities is essential to identify them and establish a therapeutic strategy to reduce or eliminate these disorders. Promising novel therapies based on the opioid imbalance (difelikefalin, a κ-opioid receptor antagonist) and the immune dysregulation theories are under evaluation 19. Even if this analysis could evaluate only the recommended pharmacotherapies reimbursed by the INHS and evaluable through administrative databases, the amount of patients with CKD-aP treated with at least a therapy recommended for CKD-aP (58.1% before and 65.1% after index date) was close to what found by Rayner and colleagues (DOPPS phase 5) in 2017 2 in hemodialysis patients with pruritus. The DOPPS phase 5 found that 68% of patients with moderate to extreme pruritus used topical treatments, 28% oral ones and the UV phototherapy was rarely prescribed. Overall, the most prescribed were antihistamines, gabapentin, sedatives and corticosteroids. Furthermore, the fact that the amount of patients treated at least once in the observational periods was not the 100% of the selected cohort deserves an explanation. Indeed, the selection was based on the presence of a CKD-aP treatment 180 days before and/or after the index date, while the free filled drug prescriptions were evaluated 365 days before or after the index date. Thus, both the algorithm and the time lapses justified the smaller amount in the observational periods.
On average, the annual direct economic impact on the INHS due to the healthcare resource consumption (i.e. reimbursed pharmaceuticals, hospitalizations and outpatient specialist services) was higher for a CKD-aP patient than for a non-CKD-aP one. These costs are quietly underestimated, mostly because administrative databases do not record the private purchase of healthcare services (for instance, self-medication through over the counter drugs for CKD-aP seems very frequent 3) and all indirect costs (e.g. those due to the loss of productivity or the caregiver support). Nevertheless, the cost analysis is crucial for understanding the possible pathway of CKD-aP patients and for comparing it with that of subjects undergoing hemodialysis but not experiencing pruritus. CKD-aP and non-CKD-aP cohorts similarly weighed to the INHS in terms of percentage distribution of the cost related to each database on the overall expenditure. Particularly, concomitant drugs corresponded with the entire pharmaceutical expenditure for the non-CKD-aP cohorts, while accounted for the 99.5% of the CKD-aP one. At the same time, the higher cost for other drugs in the CKD-aP cohort probably reflected a more unbalanced condition, which is not however possible to directly correlate with the CKD-aP itself only by means of the administrative healthcare data. Hemodialysis was predominately performed in the outpatient setting, as expected with respect to the Italian operating custom. Hemodiafiltration provided the highest cost. The performance of hemodialysis is considered critical for the successful limitation of pruritus after the therapy. This is demonstrated by the higher cost of the wide and heterogeneous hemodialysis approaches compared to that generated by the CKD-aP related treatments. The very low use of CKD-aP related therapies is, in turn, in line with the healthcare professionals’ common stepwise management assessed by Rayner and colleagues 2, that, in case of serious pruritus, tended to increase the dialysis dose before prescribing medications. Changing the hemodialytic method has been suggested as an approach to treat CKD-aP based on the toxin deposition theory, without however showing any real improvement 16. Whereas, high-flux hemodialysis, hemodiafiltration with hemoperfusion and high-permeability hemodialysis have shown significant relief of uremic pruritus compared to the conventional hemodialysis 20. Nevertheless, beyond the unclear effectiveness of the dialysis time increase 16, Sukul and colleagues 1 showed that the post dialysis recovery time from pruritus increases with the severity of the pruritus itself, contributing to add to the already high physical and psychological symptom burden and to missed treatments and withdrawal from dialysis, which was found significant among hemodialysis patients. Moreover, the potential loss of autonomy to reach the dialysis center, that inevitably causes the need for transport and increases the indirect costs related to dialysis, has been previously demonstrated in Italy 22. Our findings about costs showed plausible heterogeneous approaches to CKD-aP patients, mainly characterized by a continuous research of an effective dialysis treatment. Particularly, the not negligible use of hemodiafiltration suggested that clinicians tried an upgrade from hemodialysis to hemodiafiltration, in hopes of controlling pruritus through a better purification, probably not always successfully. Thus, the discovery of an effective treatment would lead to recommend this shift as a following step and to savings on high efficiency dialytic therapies, which are the most contributing to the average total cost of a CKD-aP patient. Although only descriptive, this cost analysis integrated the very few evidence in the CKD-aP panorama.
Strengths and limitations
Limitations of the exclusive use of this type of data are several, other than the difficulty in identifying pruritus reliably associated to CKD, previously deepened. First of all, the absence of clinical data (e.g. dialysis vintage, dialysis dose), information on renal function and other relevant patient’s characteristics could have contributed to quietly underestimate the cohort selection. The ReS database still does not link to other databases, such as dialysis registries, general practitioner’s operating system, or those collecting laboratory outcomes or the private purchase (i.e. drugs or healthcare services totally in the charge of patients). Overall, some CKD-aP patients assisted by the general medicine, together with those who do not receive healthcare by the INHS, remain undetected by this study. Moreover, the CKD identification is possible only through diagnosis and procedure codes, particularly the dialysis therapy in Italy is detectable only by means of in-hospital or outpatient procedure codes, whose precision depends on healthcare professionals. By the way, administrative data can analyse a large and unselected sample size which reliably reflects the real population. Moreover, the accuracy of administrative healthcare databases in researching CKD patients has been demonstrated high 23.