Dose adjustment in renal impairment

Editor—Vidal et al are rightly concerned that information on dose adjustment is not as well supported by evidence as our knowledge about the effectiveness of modern interventions (p 263).¹ But, as they have found, the lack of robust data to guide clinicians on the precautions to take when using drugs is woeful. Therefore, their conclusions come as no surprise to compilers of drug information.

Figure 1.

Because there are few accessible studies on dose adjustment in renal impairment, the British National Formulary (BNF) has to rely on summaries of product characteristics, which reflect data submitted for gaining marketing authorisation. However, the BNF continually adjusts its position as clinically relevant information emerges; this process is summarised in appendix 2 of Vidal et al's paper.

Vidal et al's comparison of information on drugs in renal impairment was prompted by the need to populate a computerised decision support system. On the face of it, quantitative data on dose adjustment in renal impairment seem well suited for this purpose. However, a great many factors other than renal impairment influence the choice of drug and its dose. The severity of the condition being treated, the toxicity of the drug, comorbidity, and the patient's size, age, and sex can all have a bearing on the final dose chosen, but their effect is not easily quantifiable.

And even if it were possible to quantify the full effect of the clinical and demographical information, practical constraints such as the size of the available dose form sometimes make it impossible to give the calculated dose. Therefore, the most important message is often simply that renal impairment is likely to affect the dose. The clinician would still need to titrate the dose according to the patient's clinical condition or quantitative measures—for example, international normalised ratio, blood pressure, and drug concentration for drugs such as aminoglycosides and digoxin.

The BNF takes the view that clinicians should understand the full range of a drug's clinical properties. For example, the side effects of corticosteroids include fluid retention and readers would be expected to consider this when treating a patient with renal impairment, even though the appendix on renal impairment does not make this point. The BNF also provides formulation-specific information—for example, electrolyte content—which may need to be taken into account for those with renal impairment.

Categorising the degree of renal impairment often causes difficulty, and, as Vidal et al have found, there is no universal standard. The situation is further confounded by the fact that nephrologists might be interested in characterising the overall renal physiology whereas a prescriber might be interested primarily in the efficiency of drug elimination. Often, correlation between the two is poor.

Vidal et al's paper gives further impetus to the BNF's plan to review its own advice on dosage adjustment in renal impairment. The review will include close scrutiny of the individual discrepancies that these authors have highlighted.