Chronic kidney disease (CKD) is considered one of the most serious public health problems in the world (Genovese et al., 2018; Ammirati, 2020). The CKD is classified into stages according to glomerular filtration rate (GFR) and albuminuria values with GFR values < 60 ml/min/1.73 m² indicating a CKD classified from G3a to G5, while GFR values between 60 ml/min/1.73 m² and 89 ml/min/1.73 m² indicate a CKD classified as G2 only if concomitant renal damage has been detected (KDIGO, 2013).

The prevalence of CKD is estimated as 9.1% worldwide, and, in detail, stage G3 accounts for 3.9%, stage G4 for 0.2%, and stage G5 for 0.1% (Bikbov et al., 2020). However, the number of people with CKD is expected to increase worldwide with population growth and aging. Indeed, it is widely known that kidney disease is one of the most comorbid conditions in older adults, with an overall estimated prevalence ranging from 21.4 to 47.0% and with an incidence rate of stage G4 that rises with older age (Matsushita et al., 2010; Amaral et al., 2019; Ravani et al., 2020). Aging with multimorbidity and polytherapy are the most significant factors associated with the onset and progression of end-stages of CKD but also with the hospitalization rate (Mallappallil et al., 2014; Wong et al., 2019; Schrauben et al., 2020). Moreover, the presence of CKD could be associated with several comorbidities, including hypertension, diabetes, chronic respiratory and cardiovascular disorders (Fox et al., 2012; Mahmoodi et al., 2014; Fraser et al., 2015).

Patients with at least one comorbidity are usually treated with a higher number of drugs, excreted mainly by the kidney. Older adults with GFR <60 ml/min/1.73 m² may have altered renal excreted drug pharmacokinetics and an increased risk of drug interactions which usually leads to an adjustment of their dosages to avoid toxicity (KDIGO, 2013). Older people take a broad number of medications daily, thus increasing the probability of inappropriate prescriptions (Formica et al., 2018; Roux-Marson et al., 2020). Moreover, the use of inappropriate drug prescriptions could lead to a possible worsening and progression of CKD (Schmidt et al., 2019). Several drugs, including renin-angiotensin-system (RAS)-blocking agents, anticoagulants, hypoglycemic, beta-blockers, antibiotics, analgesics, lithium, non-steroidal anti-inflammatory drugs (NSAIDs), and some antineoplastic agents are commonly known to be nephrotoxic (Taber and Pasko, 2008; Okoro and Farate, 2019; Pou et al., 2019).

The burden of CKD arouses the interest from the healthcare system for a more careful appraisal of the GFR values into the management of drug prescriptions. A recent Italian retrospective population-based study in a general practice setting showed that 56.8% of CKD affected patients received inappropriate prescriptions with a significantly greater probability when treated with more drugs, with more comorbidities and during the end stages of CKD (Barbieri et al., 2020). Nevertheless, the management of older patients with GFR <60 ml/min/1.73 m² in the hospital setting is a little-discussed issue in recent literature (Tesfaye et al., 2019; Roux-Marson et al., 2020; Troncoso-Mariño et al., 2021). Hospital admissions provide an opportunity to re-evaluate treatment regimens in patients with GFR <60 ml/min/1.73 m² and, better familiarity with nephrotoxic drugs by a physician in collaboration with a pharmacologist, could help identify appropriate preventive strategies for the management of these patients at discharge. For the reasons described above, this study aimed to evaluate the prescriptions of nephrotoxic and contraindicated drugs in older patients with GFR<60 ml/min/1.73 m² and to identify their associated factors in a real-world context.

This is the first study evaluating factors associated with contraindicated drug use in older adults with CKD in internal medicine and geriatric wards. To the best of our knowledge, few studies concerning inappropriate drug prescription in patients followed by other specialists or by general practitioners are available (Barbieri et al., 2020; Chahine, 2020; Roux-Marson et al., 2020; Troncoso-Mariño et al., 2021).

In this context, older patients with eGFR <60 ml/min/1.73 m² were mainly women: the CKD mostly affects females due to gender differences in renal pathophysiology (Cobo et al., 2016; Carrero et al., 2018). A diagnosis of CKD was registered in only 37.6% of patients even if a recent study found a formally documented diagnosis of CKD into the 59% of hospitalized older patients (Tesfaye et al., 2019). Moreover, patients with eGFR <60 ml/min/1.73 m² were particularly affected by cardiovascular disorders including hypertension, atrial fibrillation and heart failure and metabolic disease as well as diabetes. The common presence of multimorbidity in patients with CKD has been linked to a closer decline in renal function (Lee et al., 2018). Hypertension and diabetes were the most observed comorbidities in several studies (Lee et al., 2018; Chahine, 2020). It is well known that heart failure is considered the primary diagnosis at the time of hospitalization of patients with CKD and it is most related to the end-stage (Hakopian et al., 2019; Tesfaye et al., 2019; Roux-Marson et al., 2020; Schrauben et al., 2020).

The evaluation of drug used in patients with CKD showed that the number of drugs at the time of admission ranged from 4 to 8 different active substances for each patient. Several studies showed that the presence of more comorbidities could lead to polytherapy with many drugs used for each patient (Schmidt et al., 2019). Most patients were fully appropriate, and only 22.0% were in treatment with at least one inappropriate drug at admission. This data are lower than that observed in other studies, with percentages ranging from 34.1 to 90.0% (Chahine, 2020; Molnar et al., 2020; Roux-Marson et al., 2020). Moreover, remodeling of therapies during hospitalization was reported in about the 4.0% of subjects, and a reduction in the inappropriate use of medications at discharge mainly occurred in patients at the end-stage of CKD (G4 and G5). This finding agreed with the results of another study that showed a significant decrease in inappropriate drugs from admission to discharge, highlighting once again the importance of a careful assessment by the clinician (Tesfaye et al., 2019).

The recorded diagnosis of CKD did not modify the percentage of patients treated with at least one contraindicated drug at discharge, as previously noted (Barbieri et al., 2020). Specifically, the increase in the number of different drugs taken and the severity of CKD were the factors associated with an increased risk of inappropriateness. Further evidence showed that the number of several medications was the most factor of inappropriate prescription (Chang et al., 2015; Doody et al., 2015; Laville et al., 2018; Chahine, 2020). The correlation between inappropriateness and the severity of CKD was controversial. In line with other findings (Laville et al., 2018; Barbieri et al., 2020), the CKD stage was an independent factor of contraindicating drug use. Otherwise, no significant difference was noticed in the use of potentially inappropriate medications in patients in advanced CKD stages (Chahine, 2020) and patients at the CKD stage G3 were significantly more likely to receive at least one nephrotoxic drug compared to those at end-stages (Okoro and Farate, 2019).

Concerning the use of every single active substance, low-dose acetylsalicylic acid (13.3% of inappropriate drugs) was the mostly used medication at the time of admission. This consumption was reduced by −1.2% at discharge even if it was considered the first inappropriate drug also at discharge (12.2% of patients). Specifically, acetylsalicylic acid was used in these patients as a platelet aggregation inhibitor. A higher use of acetylsalicylic acid in patients with CKD was observed in other studies (Ingrasciotta et al., 2014; Abd ElHafeez et al., 2019; Okoro and Farate, 2019; Barbieri et al., 2020). Patients mostly treated with low-dose acetylsalicylic acid were those with a registered diagnosis of CKD. This could be justified by the comorbid peripheral vascular disease and ischemic heart conditions and could be explained because no valid therapeutic alternative to low-dose acetylsalicylic acid is available in clinical practice for these certain conditions. Moreover, the most recent KDIGO guidelines on the management of patients with CKD highlight the importance to use aspirin to prevent cardiovascular events and suggest to not stopped this therapy in case of diabetes or nervous and/or cardiovascular diseases, always considering the higher risk of major bleeding due to the defective platelet adhesion and aggregation, and other intrinsic platelet defects in patients with CKD (KDIGO, 2013; Gallagher et al., 2019). Other contraindicated drugs that experienced a reduction in use at discharge were diclofenac, glimepiride, metformin, and olmesartan. Reduction in their consumptions could be related to substitution with other alternative therapeutics, including glucagon-like peptide-1 receptor agonists (GLP-1 RA) and thiazolidinediones (Barone et al., 2017; Kristensen et al., 2019) or analgesics (Baker and Perazella, 2020). The NSAIDs, comprising diclofenac, should be avoided in older CKD subjects even if their consumptions continue to be frequent after discharge because of their availability over the counter (Mirishova and Hammad, 2018). Concerning angiotensin receptor blockers, guidelines recommended their use as first-line therapy in patients with CKD but may cause hyperkalemia (Ku et al., 2019; Sinha and Agarwal, 2019). However, the use of olmesartan is contraindicated when concomitantly administered with aliskiren in patients with CKD for the higher risk of adverse effects (Agenzia Italiana del Farmaco Banca Dati Farmaci, 2021). The use of metformin must always be avoided in patients with advanced CKD stage because it leads to an increased risk of lactic acidosis (Tanner et al., 2019; Molnar et al., 2020). However, the incidence of lactic acidosis associated with its use has been negligible, and guidelines are largely flexible, mainly in the mild-moderate stages of CKD (Lazarus et al., 2018). Moreover, it could be essential to pose attention to the drug-drug interactions, including the concomitant use of low-dose acetylsalicylic acid and metformin that could result into a moderate severity of type C (Santos-Díaz et al., 2020).

Another critical concern was related to the increase in the inappropriate use of fondaparinux, calcium carbonate, and antacids after hospitalization because they often did not have valid alternatives. Fondaparinux is preferred when no other choice is available including warfarin and other anticoagulant drugs (Saheb Sharif-Askari et al., 2017); however, it is associated with a higher probability of thrombocytopenia and bleeding especially in older patients for a reduced drug elimination (Agenzia Italiana del Farmaco, 2021). Calcium carbonate is used in CKD patients due to common hypocalcaemia or as a phosphate binder if alternatives cannot be found; however, in a recent study, it was shown that calcium carbonate was associated with increased vascular calcification that makes calcium-based phosphate binders not a safe option for CKD patients (Neto and Frazão, 2021); the last Clinical Practice Guideline Update for the Diagnosis, Evaluation, Prevention, and Treatment of CKD-Mineral and Bone Disorder suggested that excess exposure to calcium through diet, medications, or dialysate may be harmful across all GFR categories of CKD. Nevertheless, available evidence does not conclusively demonstrate that calcium-free agents are superior to calcium-based agents leaving the choice of possible alternative therapies (Kidney Disease: Improving Global Outcomes, 2017). Overall, the hospital doctor of the internal medicine and geriatrics wards sees quite complex patients at the time of admission, mainly older adults, in polytherapy and with multi-organ damage that could compromise the patient’s own life. For this reason, physicians prioritize the clinical stabilization of the patient for the entire duration of the hospitalization. This could overshadow some more competent evaluations from the point of view of pharmacologist and consequently neglect, at least initially, considerations regarding the pharmacokinetic and pharmacodynamic characteristics of the drugs used during hospitalization and the possible consequences of toxicity on some organs, including nephrotoxicity. At discharge, the patient’s general condition should be improved, and drug therapy may be adjusted according to the new clinical status. In some cases, no valid therapeutic alternatives with better efficacy and safety profiles than those already administered to the patient are available. Moreover, frequent monitoring of renal markers could be useful to constantly re-evaluate the risk/benefit profile of drugs highlighting the concept of personalized drug therapy. Additionally, the clinical pharmacologist involvement, a physician commonly consulted in European Hospitals due to his expertise in drug therapy and drug-drug interactions, becomes really important in the multidisciplinary management of patients with CKD. It improves the appropriate medication management not only during the hospital stay but especially at discharge for treatments to be continued at home (St. Peter et al., 2013; Alicic et al., 2016; Al Raiisi et al., 2019; Molnar et al., 2020). Moreover, the intervention of pharmacists has a positive impact not only outcomes, but also on the humanistic and economic outcomes of patients affected by CKD (Salgado et al., 2012).

The present study highlights an overall increase in therapeutic appropriateness in hospitalized older patients affected by CKD, despite a little percentage of therapeutic inappropriateness also occurred at discharge. For this reason, data could be further improved by paying greater attention to possible therapeutic alternatives. Therefore, the need for a closer collaboration between the general practitioner, the hospital physician and the clinical pharmacologist must be consolidated with the feasible aims of therapeutic appropriateness.