Does Starting Allopurinol Prolong Acute Treated Gout?
Does Starting Allopurinol Prolong Acute Treated Gout?
Traditionally, allopurinol is not initiated during an acute gout attack to avoid prolonging the painful gouty attack. The 2012 ACR Guidelines for the Management of Gout suggest that urate-lowering therapy can be started during an acute gout attack provided effective acute management is instituted. These recommendations were based on "consensus opinion of experts, case studies, or standard of care." The results of our study support the 2012 ACR Guidelines for the Management of Gout.
We conducted a placebo-controlled, double-blind study of allopurinol initiation in patients receiving treatment for acute gout attacks. All patients met standard indications for beginning pharmacologic urate-lowering therapy. Prophylactic colchicine or meloxicam was given upon initiation of the study drug for the entire study. We developed a stringent definition of acute inflammatory gout resolution incorporating patient-reported improvement in pain combined with physical examination measures of swelling, warmth, and tenderness. This definition was based on our clinical experience, as there is no uniform definition of acute gout attack resolution. Patient-rated pain is subjective and varies significantly between individuals. Therefore, we included objective physical examination measures in addition to patient-rated pain and noted the difference in pain within individuals as well. We believe our criteria defined clinical resolution, a point at which the patient's pain was minimal or resolved and no longer affecting their functional status. Secondary end points show that the most acute symptoms from gout resolved well before patients met our study criteria for resolution, which included even minor symptoms at the end of the attack.
The acute gout attack resolved by study criteria after an average of 13.4 days in subjects taking placebo and 15.4 days in subjects taking allopurinol. The difference between groups was not statistically significant. Furthermore, the majority of the patients in both groups displayed very low PGA scores and patient-rated pain scores by days 3 to 4 (visit 2). Even this early into the study, the mean pain scores for all patients were less than 3. This indicates that the difference in days to gout attack resolution was not statistically significant, nor was it clinically meaningful from that point on, although there could have been a difference on days 2 and 3.
Allopurinol initiated at a low dose, and gradually increased, was well tolerated, with 1 serious adverse event reported. One subject experienced recurrent epistaxis requiring nasal packing. This patient was also taking warfarin for the treatment of atrial fibrillation, and his coagulation profile remained in the therapeutic range. Epistaxis associated with allopurinol is reported in the package insert, although a literature search failed to find any case reports of this association.
Taylor et al reported the only other study of allopurinol initiation in acute gout. Similar to our study, this was a double-blind, placebo-controlled investigation of 57 patients. However, there were several differences in methods with this study. Taylor et al did not require standard indications for initiation of allopurinol therapy, and their patients were excluded if they had tophi, history of congestive heart failure, or anticoagulant use. We allowed the primary physician to determine the appropriate treatment modality for the acute gout attack, whereas Taylor et al used a standardized dose of indomethacin and prophylactic dose of colchicine. Our primary end points differed as well. We defined DTR of the gout attack as our primary end point, whereas Taylor et al measured pain at all days from days 1 to 10 by visual analog scale and self-reported gout flares between days 1 and 30. Finally, in the study by Taylor et al, allopurinol was initiated at 300 mg daily, rather than in the stepped increments recommended in the 2012 ACR guidelines. However, despite the differences in methods and dosing, Taylor et al similarly found that allopurinol initiation during an acute gout attack caused no significant difference in daily pain on days 2 to 10 (P = 0.62) or recurrent flares (P = 0.61).
Limitations of our study include a nonuniform approach to treating the acute gout attack, allowing the treating physician to determine the appropriate treatment based on the severity of the attack, joints involved, and patient preference. This was intended to mimic real-world practice and to give the patient the greatest opportunity for improvement. A large proportion of our patients received corticosteroids as initial treatment for acute gout, which may reflect local practice patterns. The protocol called for prophylaxis with colchicine or NSAIDs, according to common practice and the recommendations of the ACR. The acute treatment methods did not include colchicine alone. The treatment methods and prophylaxis used for patients in this study may have been the reason that they did not suffer gout flares with allopurinol initiation. Furthermore, exclusion criteria limited patients with certain comorbidities that limit the use of acute gout prophylaxis, including elevated liver enzymes and reduced GFR. Therefore, the results of our study may not apply to these patient populations. We started allopurinol at 100 mg daily, which is in accordance with recommendations of the ACR. Our results may not apply to patients initiating allopurinol at higher doses. Although we report early uric acid level measurements, the goal of our study was not to achieve a steady-state uric acid level of less than 6 mg/dL, but rather to test the effect of allopurinol initiation during the acute treatment phase of gout. Our study design allowed for a 28-day follow-up and does not address the risk for recurrent attacks beyond this period.
Our post hoc power analysis suggested that a larger number of patients would need to be studied to detect statistical significance for the difference in time to resolution of gout between arms of this study. Post hoc power analyses can produce a biased estimate of true power. Nonetheless, a larger study, using our data for a new prestudy power analysis, should be done to confirm our results.
Finally, although the investigators felt that study blinding was effective, their "guess" of each patient's randomization prior to unblinding was correct 85% of the time. This occurred despite additional steps taken, to include blinding each patient's investigator to the monitoring laboratory results after an interim analysis. This suggests there could have been unblinding during the study. If this did occur, however, it was not passed along to the patients, who were the source of the secondary data, including the pain scores, which were truly blinded and corroborated the primary end point.
Our results clearly show that the severe symptoms of gout resolved quickly in both arms of the study, and therefore our protocol captured data that included even minor residual symptoms. If the severe symptoms of the acute gout attack were prolonged by even 2 days, then we would conclude that allopurinol should not be started during treatment of an acute attack. In order to maximally challenge our statistical conclusion, our intent-to-treat analysis intentionally favored faster resolution in the placebo arm, but even then the difference was not statistically significant. Therefore, we are confident that if a clinically meaningful prolongation of gout were caused by the initiation of allopurinol, we would have detected it in this study. Our study documented gout resolution as of the dates of the visits. Therefore, a gout attack could have resolved between visits, resulting in overestimation of the time to resolution. However, the same method was used for both study arms, and definition of gout resolution was intentionally based on the objective criteria available at in-person visits.
Our study was not designed to determine the benefits, if any, of initiating allopurinol concomitant with treatment for an acute attack. However, doing so has the potential to decrease health care costs. Inadequate treatment of gout in addition to poor patient adherence to urate-lowering therapy has long been recognized. Initiating prophylaxis concomitantly with treatment for the acute gout attack may prevent delays in the start of treatment because of noncompliance and reduce the number of follow-up visits needed and the lifetime number of acute gout attacks requiring treatment. Based on our results and clinical experience, we advocate starting allopurinol only if the patient meets the standard indications for beginning urate-lowering therapy and with the dose titration as recommended in the ACR 2012 guidelines.
In summary, we found that initiating allopurinol during an acute gout attack does not adversely affect the resolution of the acute, treated attack. In addition, there are now 2 differently designed randomized controlled trials that support the expert recommendation that urate-lowering therapy can be started during an acute attack provided effective acute management is instituted, as stated in the 2012 ACR Guidelines for the Management of Gout.
Discussion
Traditionally, allopurinol is not initiated during an acute gout attack to avoid prolonging the painful gouty attack. The 2012 ACR Guidelines for the Management of Gout suggest that urate-lowering therapy can be started during an acute gout attack provided effective acute management is instituted. These recommendations were based on "consensus opinion of experts, case studies, or standard of care." The results of our study support the 2012 ACR Guidelines for the Management of Gout.
We conducted a placebo-controlled, double-blind study of allopurinol initiation in patients receiving treatment for acute gout attacks. All patients met standard indications for beginning pharmacologic urate-lowering therapy. Prophylactic colchicine or meloxicam was given upon initiation of the study drug for the entire study. We developed a stringent definition of acute inflammatory gout resolution incorporating patient-reported improvement in pain combined with physical examination measures of swelling, warmth, and tenderness. This definition was based on our clinical experience, as there is no uniform definition of acute gout attack resolution. Patient-rated pain is subjective and varies significantly between individuals. Therefore, we included objective physical examination measures in addition to patient-rated pain and noted the difference in pain within individuals as well. We believe our criteria defined clinical resolution, a point at which the patient's pain was minimal or resolved and no longer affecting their functional status. Secondary end points show that the most acute symptoms from gout resolved well before patients met our study criteria for resolution, which included even minor symptoms at the end of the attack.
The acute gout attack resolved by study criteria after an average of 13.4 days in subjects taking placebo and 15.4 days in subjects taking allopurinol. The difference between groups was not statistically significant. Furthermore, the majority of the patients in both groups displayed very low PGA scores and patient-rated pain scores by days 3 to 4 (visit 2). Even this early into the study, the mean pain scores for all patients were less than 3. This indicates that the difference in days to gout attack resolution was not statistically significant, nor was it clinically meaningful from that point on, although there could have been a difference on days 2 and 3.
Allopurinol initiated at a low dose, and gradually increased, was well tolerated, with 1 serious adverse event reported. One subject experienced recurrent epistaxis requiring nasal packing. This patient was also taking warfarin for the treatment of atrial fibrillation, and his coagulation profile remained in the therapeutic range. Epistaxis associated with allopurinol is reported in the package insert, although a literature search failed to find any case reports of this association.
Taylor et al reported the only other study of allopurinol initiation in acute gout. Similar to our study, this was a double-blind, placebo-controlled investigation of 57 patients. However, there were several differences in methods with this study. Taylor et al did not require standard indications for initiation of allopurinol therapy, and their patients were excluded if they had tophi, history of congestive heart failure, or anticoagulant use. We allowed the primary physician to determine the appropriate treatment modality for the acute gout attack, whereas Taylor et al used a standardized dose of indomethacin and prophylactic dose of colchicine. Our primary end points differed as well. We defined DTR of the gout attack as our primary end point, whereas Taylor et al measured pain at all days from days 1 to 10 by visual analog scale and self-reported gout flares between days 1 and 30. Finally, in the study by Taylor et al, allopurinol was initiated at 300 mg daily, rather than in the stepped increments recommended in the 2012 ACR guidelines. However, despite the differences in methods and dosing, Taylor et al similarly found that allopurinol initiation during an acute gout attack caused no significant difference in daily pain on days 2 to 10 (P = 0.62) or recurrent flares (P = 0.61).
Limitations of our study include a nonuniform approach to treating the acute gout attack, allowing the treating physician to determine the appropriate treatment based on the severity of the attack, joints involved, and patient preference. This was intended to mimic real-world practice and to give the patient the greatest opportunity for improvement. A large proportion of our patients received corticosteroids as initial treatment for acute gout, which may reflect local practice patterns. The protocol called for prophylaxis with colchicine or NSAIDs, according to common practice and the recommendations of the ACR. The acute treatment methods did not include colchicine alone. The treatment methods and prophylaxis used for patients in this study may have been the reason that they did not suffer gout flares with allopurinol initiation. Furthermore, exclusion criteria limited patients with certain comorbidities that limit the use of acute gout prophylaxis, including elevated liver enzymes and reduced GFR. Therefore, the results of our study may not apply to these patient populations. We started allopurinol at 100 mg daily, which is in accordance with recommendations of the ACR. Our results may not apply to patients initiating allopurinol at higher doses. Although we report early uric acid level measurements, the goal of our study was not to achieve a steady-state uric acid level of less than 6 mg/dL, but rather to test the effect of allopurinol initiation during the acute treatment phase of gout. Our study design allowed for a 28-day follow-up and does not address the risk for recurrent attacks beyond this period.
Our post hoc power analysis suggested that a larger number of patients would need to be studied to detect statistical significance for the difference in time to resolution of gout between arms of this study. Post hoc power analyses can produce a biased estimate of true power. Nonetheless, a larger study, using our data for a new prestudy power analysis, should be done to confirm our results.
Finally, although the investigators felt that study blinding was effective, their "guess" of each patient's randomization prior to unblinding was correct 85% of the time. This occurred despite additional steps taken, to include blinding each patient's investigator to the monitoring laboratory results after an interim analysis. This suggests there could have been unblinding during the study. If this did occur, however, it was not passed along to the patients, who were the source of the secondary data, including the pain scores, which were truly blinded and corroborated the primary end point.
Our results clearly show that the severe symptoms of gout resolved quickly in both arms of the study, and therefore our protocol captured data that included even minor residual symptoms. If the severe symptoms of the acute gout attack were prolonged by even 2 days, then we would conclude that allopurinol should not be started during treatment of an acute attack. In order to maximally challenge our statistical conclusion, our intent-to-treat analysis intentionally favored faster resolution in the placebo arm, but even then the difference was not statistically significant. Therefore, we are confident that if a clinically meaningful prolongation of gout were caused by the initiation of allopurinol, we would have detected it in this study. Our study documented gout resolution as of the dates of the visits. Therefore, a gout attack could have resolved between visits, resulting in overestimation of the time to resolution. However, the same method was used for both study arms, and definition of gout resolution was intentionally based on the objective criteria available at in-person visits.
Our study was not designed to determine the benefits, if any, of initiating allopurinol concomitant with treatment for an acute attack. However, doing so has the potential to decrease health care costs. Inadequate treatment of gout in addition to poor patient adherence to urate-lowering therapy has long been recognized. Initiating prophylaxis concomitantly with treatment for the acute gout attack may prevent delays in the start of treatment because of noncompliance and reduce the number of follow-up visits needed and the lifetime number of acute gout attacks requiring treatment. Based on our results and clinical experience, we advocate starting allopurinol only if the patient meets the standard indications for beginning urate-lowering therapy and with the dose titration as recommended in the ACR 2012 guidelines.
In summary, we found that initiating allopurinol during an acute gout attack does not adversely affect the resolution of the acute, treated attack. In addition, there are now 2 differently designed randomized controlled trials that support the expert recommendation that urate-lowering therapy can be started during an acute attack provided effective acute management is instituted, as stated in the 2012 ACR Guidelines for the Management of Gout.
Source...