Key Findings and SoE

We found a large number of research studies about gout, yet only a relatively modest number of these studies provided evidence for some of our KQs, particularly for the treatment of acute gout: only a single placebo-controlled trial of NSAIDs for acute gout pain, two placebo controlled RCTs of colchicine, and none at all for corticosteroids or ACTH. Nevertheless, we were able to reach strong conclusions about the usefulness of these drugs because of some specific features of gout: Symptoms result from an inflammatory reaction to the deposition of urate crystals, which occurs when serum urate rises above its saturation point in the blood. Hence, in an era that predated the widespread practice of placebo-controlled trial testing of therapies, medications aimed at blocking the inflammatory response were tried as treatments. Steroids are one of the most powerful and effective anti-inflammatory medications available. Although no placebo-controlled RCTs have tested their use in acute gout, steroids have proven efficacy in other inflammatory conditions, which gives us confidence that they are effective in treating the inflammatory reaction in acute gout. At this point, a placebo-controlled trial of steroids in acute gout may well be unethical, as it would mean withholding therapies of known effectiveness (e.g., colchicine) from the placebo-treated group. Indeed, a recent, high profile study of the use of steroids in acute gout compared their use not to placebo, but to NSAIDs. Because NSAIDs also have no conclusive placebo-controlled trial evidence of their effectiveness in acute gout, could it be that this RCT, which found only minor differences in outcomes between the two treatments, actually was comparing two treatments that were equally ineffective? We think not. We believe that both drugs are effective in treating acute gout, and therefore judged the SoE as high that their use relieves symptoms by a clinically important amount—despite the lack of placebo-controlled RCT evidence.

With regard to chronic gout, we similarly used evidence from a number of sources to reach conclusions about the effectiveness of ULT at reducing the risk of acute gout attacks over time, despite the fact that this outcome has not been studied in any placebo-controlled trial of longer than a few months. We based our moderate SoE rating on the high strength evidence that ULT reduces serum urate, that serum urate level is a strong predictor of the risk of acute gout attacks, and that the open-label extension studies of randomized controlled trials of ULT have shown a graded relationship between the serum urate level achieved and the risk of acute gout attacks. We thus concluded that over time, possibly by 1 year from initiation of therapy, ULT reduces the risk of acute gout attacks. We also concluded, based on a comparison of the timing of the occurrence of acute gout attacks in the weeks following initiation of ULT, that longer courses of prophylactic treatment with colchicine or NSAIDs (greater than 8 weeks) are more effective than courses of treatment with durations of 8 weeks or less, since in the one RCT of urate lowering therapy where prophylactic colchicine or NSAIDs were continued for 6 months, no “spike” in acute gout attacks coincided temporally with the discontinuation of the prophylactic therapy, like that seen in other RCTs where prophylaxis was stopped at 8 weeks.

A third key finding of our review is that there is scant direct evidence about how much ULT to give (e.g. the concept of treating-to-a-target) and for how long to give it (are there any criteria about when ULT can be stopped, or if once started is treatment needed for life?).

The key findings and SoE are in Table 20.

Table 20

Summary of prior knowledge, findings from the systematic review, and strength of evidence, by KQ.

Findings in Relationship to What is Already Known

In general, our findings support the results of existing SRs. We did find a number of RCTs not included in prior reviews. Some of these studies were “first-of-their-kind,” such as those testing a specific dietary therapy and the duration of colchicine prophylaxis. However, most new studies either confirmed prior knowledge, or, in the case of studies of novel treatments, were not sufficiently high quality for us to draw conclusions.

Applicability

Of the 115 studies assessed in detail (not counting SRs), only 9 studies explicitly stated that patients came only from, or the study included patients from, primary care sites (including the ED and urgent care settings). Furthermore, it is likely that patients enrolled in clinical trials differ from those commonly seen in primary care settings. In the major trials of ULT, the proportion of patients with tophi is greater than 20 percent^118-121 whereas in a trial that explicitly enrolled patients from primary care, the proportion of patients with tophi was 10 percent. A population-based study of more than 50,000 gout patients in English primary care practices reported the proportion with tophi as 0.5 percent¹⁹⁴ Patients enrolled in clinical trials usually have fewer comorbidities than those seen in practice, because clinical trials have exclusion criteria. Thus, in most trials, enrolled patients probably had more advanced gout, but were better on average with respect to their other health conditions, than patients typically seen in primary care settings. We thus judged this evidence of moderate applicability to primary care.

Implications for Clinical and Policy Decisionmaking

The implications of this review for clinical decision-making follow from the identification of which interventions for gout management have evidence of an effect on clinical outcomes, either directly or through a strong indirect evidence chain. Thus, the results in Table B will be useful in policy decision-making and in the development of practice guidelines.

Limitations of the Comparative Effectiveness Review Process

For many of the KQs of interest, data were not reported on the subgroups or outcomes of interest as specified in the KQs and analytic frameworks, limiting the comparative effectiveness review. For the portion of the review on Traditional Chinese Medicine (TCM), the variability in tested interventions made comparisons across studies not justified.

Limitations of the Evidence Base

The lack of studies specifically stating that they enrolled patients in primary care settings is a limitation, as is the lack of randomized controlled studies assessing clinical outcomes for ULT (such as recurrent acute gout flare after 1 year) and intervention studies of dietary therapies for management of chronic gout. Longer term studies will be needed to assess the degree to which ULT reduces acute gout attacks relative to the adverse events of long term use of the available medications.

Research Gaps

The concept of “treat-to-target” (TTT) in gout, while supported by indirect evidence, has been untested. Guidelines and recommendations about TTT thresholds already vary, for example, < 6mg/dL for all gout patients versus < 5mg/dL for patients with significant gout morbidity. However, for many gout patients in primary care practice whose gout is well controlled on ULT, no data support such targets. In fact, the results of one cohort study suggest that once gout has been asymptomatic for 5 years, ULT might be discontinued for many years (as long as serum urate levels remain acceptable, e.g., < 7mg/dL).¹⁹⁰ Therefore, the most important research gap is a RCT comparing different TTT levels in patients with gout and elevated serum urate.

Treatment decisions are likely to be preference-sensitive, and studies are needed to assess patient preferences for different outcomes (for example, to what degree do patient preferences differ for outcomes such as a decrease in risk from 2 percent to 0.5 percent for an acute gout attack in the coming year versus a 5 percent chance of a skin rash and a less than 1 percent chance of a very serious skin rash).

Likewise, in spite of the many observational studies linking dietary factors with risk for gout, few studies have assessed the effect of specific dietary advice. Some dietary advice, such as generic advice to lose weight in overweight and obese patients, has evidence of benefit for other conditions and can be advocated in gout patients without additional data (e.g., it is always indicated to recommend dietary weight loss in patients who are obese). But primary care providers could more confidently recommend gout-specific dietary advice if compelling evidence supported an effect of such dietary changes on the risk for gout attacks or other gout-related outcomes. Therefore, another important research gap is evidence from RCTs for specific dietary changes (such as reducing or eliminating sugar-sweetened beverages or high-fructose foods, adequate hydration, restriction of alcohol, increase in low fat dairy consumption, and even restriction of high purine foods) compared with standard healthy diet advice and weight loss in reducing the risk of gout attacks.

A third research gap is the better characterization of adverse events from ULT and how they may be minimized. If the rare but serious adverse events from ULT could be further minimized, for example by HLA typing for predisposition, then the benefit/risk profile of ULT would further improve and make more patients eligible for treatment.

An additional research gap concerns prophylaxis when initiating ULT therapy. The optimal duration of such therapy has not been experimentally tested, and the comparative benefits/risks of all three agents used for acute attacks (colchicine, NSAIDs, oral steroids) have not been established.

Conclusions

Several drugs show moderate-to-high evidence of benefit in terms of reducing pain in patients with acute gout. It is clear that urate lowering therapy achieves its goal of lowering urate levels. Decreased serum urate should lead, over time, to a reduction in gout attacks, but the benefits and harms of long term urate lowering therapy have yet to be demonstrated directly. Patient preferences are likely to be important in decision-making (as specified above), and having better estimates of the size of the benefit of urate lowering therapy will make clinicians and patients more knowledgeable about the risk: benefit trade-off for the different decisions.