Evaluation of the tolerance of ceftriaxone by subcutaneous route in patients ≥75 years old in geriatric departments: a prospective observational study

Abstract

Objectives

The subcutaneous (SC) route provides an alternative to the IV or oral route for drug administration in the elderly. The benefits of SC administration have been proven for hydration but are still debated for antibiotics because tolerance remains uncertain, especially in the frail geriatric population. Here, we aimed to improve current knowledge concerning the tolerance profile of ceftriaxone SC administration at both the systemic and cutaneous level, as well as in terms of pain.

Patients and methods

This was a prospective descriptive study of SC ceftriaxone tolerance in a geriatric department. We included all patients over 75 years of age who received a prescription for SC ceftriaxone in our hospital over a 5 month period.

Methods

We evaluated the systemic and local tolerance of SC ceftriaxone. Nurses were asked about their perceptions concerning its use.

Results

Among 117 patients, 57% presented with pain and 60% with a mild local adverse effect, such as the formation of oedema in one-third of patients, induration or transient erythema. Finally, there were no serious local adverse effects and two systemic adverse effects were observed (one diarrhoea and one Clostridioides difficile colitis). Pain was mainly related to the skin breach and oedema formation.

Conclusions

We did not find any worrying signs concerning the use of SC ceftriaxone but this study shows that its wide use must consider pain management, which is often overlooked.

Introduction

The subcutaneous (SC) route is an alternative to the IV or oral route for drug administration in the elderly because of the precarious venous capital, behavioural or swallowing disorders. Its benefits have been proven for hydration^1,2 but are still debated for antibiotics such as ceftriaxone,³ especially since the European restriction in 2014 and that of the French National Agency for the Safety of Medicines and Health Products in 2021, pointing out the lack of data on its efficacy and the risk of skin necrosis.⁴ Despite ‘Dear Doctor’ letter release, 96% of geriatricians and infectiologists still carry out SC administration of antibiotics in clinical practice.⁵ Nevertheless, tolerance to local adverse effects (LAEs) and induced pain remain uncertain in the geriatric population.⁶ We aimed to improve current knowledge concerning the tolerance profile of ceftriaxone SC administration in geriatric departments.

Patients and methods

Study population and data collection

This was a prospective, observational and monocentric study at the Amiens Picardie university hospital center. All patients over the age of 75 in geriatric medicine (57 patients), rehabilitation (40 patients) and nursing homes (190 patients) were included if they had received ceftriaxone SC over the period from 22 March 2021 to 15 August 2021. At each administration, the nurse noted the injection site, the condition of the skin and a scale, validated in geriatrics, evaluated the pain (Algoplus scale) according to the time of injection (before, during and after the infusion).⁷ Data on ceftriaxone preparation and administration practices (dilution volume, solvent, needle used, time of infusion) were collected. Adverse effects (AEs) were reported on dedicated questionnaires. To avoid missing patients, our clinical pharmacist checked new ceftriaxone prescriptions on the service every morning. Medical data, geriatric parameters and treatments administered were obtained from electronic clinical records and questionnaires completed by nurses then collected anonymously in an Excel^® database. Patients who received ceftriaxone SC with a follow-up of less than 2 days were excluded.

Ethics

The local ethics committee approved the study (no. PI2021_843_0042). Due to the study’s observational design, formal consent by patients (or legal guardians) was not required. Each AE occurring was reported to the regional pharmacovigilance centre.

Statistical analysis

All values are presented as means or medians, with standard deviation. The normality of the variable distribution was defined visually or by a Shapiro test. The χ² test was used to compare the frequencies and the Student’s t-test to compare the means (or by non-parametric tests if necessary). P values of <0.05 were considered statistically significant. All statistical analyses were performed using R^® software (R i386.4.0.2).

Results

We included 117 patients and 20 patients were excluded because they had received less than 2 days of treatment. The average age was 88 years and most were women (65%). The mean ± SD Charlson score was 8 ± 2. Among comorbidities, neurocognitive disorders, arterial hypertension and renal failure were the most frequent (72%, 72% and 54%, respectively). Regarding activities of daily living (ADL) levels, there were 3% independent patients (ADL 6/6), 42% with moderate dependency (ADL from 3 to 5) and 55% with severe dependency (ADL from 0 to 2). Swallowing or behavioural problems concerned 45% of patients. Patient characteristics are presented in Table 1.

Prescription of ceftriaxone

In 58% of cases, ceftriaxone was prescribed by SC as first-line therapy, given the difficulty of taking orally or to promote mobilization in patients. For the others, it followed IV antibiotic therapy (32% ceftriaxone, 6% amoxicillin/clavulanic acid and 4% cefotaxime) with a median relay performed on the third day. All injections were made at a dose of 1 g. SC ceftriaxone was prescribed for an average of 9 ± 3 days for urinary tract infection (63%), pneumopathy (40%), bacteraemia (7%) or digestive infection (4%). In 85.5% of cases, antibiotic therapy was initiated probabilistically. The infection could be documented for 66 patients (56%). A total of 19 species were identified, mainly Escherichia coli (n = 41) and Proteus mirabilis (n = 8) in urine, and Klebsiella pneumoniae (n = 10) in urine and in a lung sample. There were 10 organisms resistant to ceftriaxone (Table 2).

Preparation and administration of ceftriaxone

We interviewed 85 nurses, of whom 52 (61%) responded. For ceftriaxone infusions, they mainly used soft polyurethane (94%) 24-gauge (65%) catheters, with 0.9% sodium chloride as solvent (85%), of which the volume varied from 50 to 100 mL, and for a duration of 30 min (80%) controlled by a volumetric infusion pump (Agilia^® VP). The subcutaneous catheter was changed daily, with the exception of three nurses, who could leave it in place for up to 4 days.

Pain assessment

Among the 117 patients, 67 (57%) presented with pain during at least one of the treatment administration times, including 50 (43%) during the introduction of the catheter, 36 (31%) during the infusion and 9 (8%) after infusion. There were 28 (24%) patients who felt pain exclusively during the introduction of the catheter, 11 (9%) exclusively during the infusion, and 3 (3%) exclusively after the infusion. Among characteristics of the patients (Table 1), the concomitant treatments of the class of analgesics and those taken subcutaneously (SC hydration and low-molecular-weight heparins), only SC hydration was associated with pain. There was no association between pain and injection site, which was 74.5% on the thighs, 23.1% on the abdomen and 2.3% on the back.

Assessment of AEs

The injection was performed on healthy skin in 86.4% of cases but the infused area was already oedematous in 10.8% of cases, showed redness in 2.3%, or already showed induration in 1.3%. There were two systemic AEs observed, one simple diarrhoea and one diarrhoea with Clostridioides difficile colitis. In total, 44 patients (37.6%) experienced at least one LAE and a total of 51 LAEs were identified. We observed no cases of necrosis, collection or infection. There were 36 cases of local oedema (30.8% of patients) for a few hours, 8 of local induration (6.8% of patients) and 7 of local erythema (6% of patients). No patient had more than two LAEs. There was no association between the presence of oedema before injection and pain but the formation of oedema and the appearance of cutaneous induration were associated with pain. The occurrence of an LAE was not significantly associated with patient characteristics, concurrent administration of other SC treatments, antiplatelet or anticoagulant therapy, or duration of treatment.

Discussion

In this cohort of geriatric patients treated with 1 g SC ceftriaxone, 60% had no LAEs, and no serious LAEs were observed. Between 1 January 1985 and 1 January 2022, the French national pharmacovigilance database included 10 cases of SC necrosis with ceftriaxone and LAEs were mainly found at doses of 2 g (EUCAST recommendations), i.e. higher than those usually used in French geriatric services.^8,9 However, there is under-reporting bias in the pharmacovigilance database and reported effects are limited. We observed few of these LAEs, possibly due to the use of a 1 g dose, nurses used to this practice, and a controlled infusion rate of approximately 30 min, unlike other studies where the infusion was sometimes rapid (<5 min) and found 8% haematomas and 7% indurations.⁶ A recent systematic review recalled the good tolerance of the SC administration of several antibiotics including ceftriaxone, which supports our comments.¹⁰ However, in our study it is possible that certain minor effects were not noted, and that certain systemic effects, particularly late ones, were not reported. The high incidence of pain in our study (57%), unlike others evaluating it at 12%, is probably explained by the fact that we systematically sought it.^7–9 Several techniques could be used to reduce the pain, like the use of the same catheter by keeping it in place for up to 3 days when possible, which would avoid daily skin breach. Furthermore, the use of a local anaesthetic (transcutaneous patch, cream) has proven its effectiveness before the placement of venous access, and before invasive procedures (blood gas, lumbar puncture).^11–13 The application of local heat would also make it possible to limit the formation of oedema due to its vasodilating effect, which accelerates the rate of absorption. Some palliative care units already use this technique before placing SC access or injecting heparin. Finally, doctors and nurses should consider spacing out the time between SC treatments and reserving an ‘exclusive’ infusion site for antibiotic therapy. By analogy with SC hydratation,¹⁴ the use of SC ceftriaxone in elderly patients should make it possible to offer an alternative in the event of poor venous access, reduce confusion when there is cognitive disorder, promote early functional recovery and reduce the length of stay.^14–16 However, pharmacology data specific to SC ceftriaxone in elderly are still lacking to date.

Conclusions and implications

In conclusion, the use of ceftriaxone by the SC route can be considered as an alternative to the IV route in the management of geriatric patients even with cognitive disorder. Local pain and AE are frequent but mild and transient. They could be reduced by non-drug measures and the use of local analgesics. However, it must be kept in mind that systemic AEs can still occur, as with the IV route.

Funding

This study was carried out as part of our routine work.

Transparency declarations

The authors declare that they have no conflicts of interest.

References