Renal Toxicity Associated with Salsalate in Elderly Patients with Anemia

We are writing to caution the geriatrics community about the possibility of serious adverse renal toxicity occurring with the use of salsalate in older individuals. Salsalate, a non-steroidal anti-inflammatory, non-acetylated form of salicylate, is one of the oldest drugs available to clinicians, and has been used to treat inflammatory conditions such as the rheumatologic disorders. (1) Though not as widely used as aspirin (the acetylated form), it has had a recent renaissance in use, in part because of its relative lack of COX-2 inhibition and resultant fewer gastrointestinal (GI) side effects(1,2). In particular, because of its ability to regulate the NF-κB pathway (3), it has been studied in a number of recent clinical trials using doses of 3.0 – 4.5 gm/day for the management of type 2 diabetes (1,4). It has shown the potential to reduce insulin resistance and other risk factors for diabetes and has been said to do so with “few if any side effects” (1,5,6). However these trials have included relatively few elderly patients. For example in the Goldfine trial the mean age of subjects was 56 (2). This trial did note a number of subjects, greater in the salsalate than the placebo arm, who had mild decreases in renal function, but no serious adverse renal toxicities were noted.

Recently as part of the work of an NIA supported consortium, The Partnership for Anemia: Clinical and Translation Trials in the Elderly (PACTTE) (7), we initiated a Phase II trial of salsalate to treat Unexplained Anemia of the Elderly (UAE) (www.clinicaltrials.gov, NCT01506726) based on the hypothesis that an anti-inflammatory agent could improve hemoglobin level because of its impact on the putative contribution of chronic age related inflammation to UAE. We chose salsalate because of its reputed advantageous side effect profile. The target enrollment was 32 in each group. We randomized subjects ≥65 years old with UAE, Interleukin-6 levels >1.0 pg/ml, with 6-minute walk times below the expected range for median age, and eGFR > 30ml/min, (4 variable MDRD),1:1 to salsalate or placebo, with hemoglobin levels and walking speed at 6 months as the primary outcome measures. The initial dose was 1.5gm/day; if tolerated, escalating to 3.0 gm/day.

As has been the case for other trials of UAE (7), we found it very difficult to recruit to this study. This was further complicated by the observation of adverse events early on in the study that was ultimately terminated early on advice of the DSMB, after enrollment of only 11 total subjects. One subject (baseline creatinine 1.3mg/dL) treated with the initial dose for two weeks, had diarrhea associated with nausea, abdominal pain, dehydration, and acute renal failure (creatinine 5.7mg/dL), which partially resolved after vigorous hydration. Another subject (baseline creatinine 0.8mg/dL) had diarrhea and a decrease in renal function (peak creatinine 1.5mg/dL) after being treated with the initial dose for 1 month. As a result the patient was taken off the drug. In this subject, the diarrhea recurred following re-administration; therefore the drug was discontinued again, and the patients received intravenous fluids, with no recurrence of the renal dysfunction. Though diarrhea (1 of 5) and nausea (2 of 5) were also seen in the placebo group, no decreases in renal function were noted. Attributable adverse events are shown in Table 1.

Table 1.

Attributable Adverse Events

Events	Salsalate		Placebo
	# Subjects 6		# Subjects 5
	NS	S	NS	S
Gastrointestinal (GI) (Nausea, Vomiting, Diarrhea)	3	1	3	0
Musculoskeletal (Arthralgia, Stiffness)	3	0	2	0
Malaise	2	0	2	0
Renal Function Impairment	1	1	0	0
Weight Loss	0	0	2	0
Dehydration	1	0	0	0
Hypoglycemia	1	0	0	0
Hypoacusis	0	1	0	0

S – Serious (Hospitalization or important medical event)

NS - Non Serious

While the number of subjects reported here is obviously too small to draw definitive conclusions, we believe that our experience suggests that if salsalate is to be used in older individuals, especially those who may be frail or pre-frail, practitioners should be especially alert to the possibility of renal failure especially in the context of diarrhea and dehydration. Additional concern is that these adverse events occurred in our subjects at a dose half that used in the reported trials. Given that older persons are frequently underrepresented in clinical trials, our observations help highlight the importance of having sufficient numbers of older subjects in trials of agents likely to be used in that population.