Cefazolin-Associated INR Elevation: A Case Report

Abstract

Purpose: To describe a case of significantly elevated international normalized ratio (INR) in a patient on apixaban receiving treatment with intravenous cefazolin in the setting of coronavirus disease 2019 (COVID-19) infection and malnutrition. Summary: A 74-year-old male patient on apixaban receiving cefazolin for osteomyelitis in the setting of COVID-19 and poor nutritional intake presented with internal jugular tunneled catheter site bleeding and an INR of greater than 22.5. Laboratory abnormalities and bleeding concerns were successfully managed with vitamin K and changing antimicrobial therapy from cefazolin to daptomycin. Follow-up labs one week later demonstrated a sustained improvement in coagulopathy. Conclusion: INR prolongation believed to be secondary to cefazolin can be effectively managed with administration of vitamin K and conversion of antimicrobial therapy to an alternative agent.

Introduction

Mechanisms of clot formation can be disrupted by multiple medical conditions and therapeutic interventions. Cefazolin is a common intravenous first-generation cephalosporin used for a variety of infections. Cephalosporins, including cefazolin, are known to interact with vitamin K antagonist anticoagulants through several proposed mechanisms. ¹ Published case reports and related research have also associated cefazolin use with coagulation abnormalities, including hypoprothrombinemia independent of concurrent anticoagulation.^2-8 Less is known about the interaction between cefazolin and direct oral anticoagulants (DOACs), including any subsequent coagulation disturbance or bleeding risk. Finally, acute medical comorbidities like coronavirus disease 2019 (COVID-19) and disseminated intravascular coagulopathy (DIC) can independently contribute to hypercoagulability and clotting factor consumption, while a patient’s overall health status or concurrent medications may impact the activity or production of clotting factors.^9-19 When these conditions and therapeutic interventions cooccur, the resulting impact on mechanisms of coagulation may be significant.

Here, we report a case of a 74-year-old male patient receiving concurrent apixaban and cefazolin who developed a significant prolongation of prothrombin time/international normalized ratio (PT/INR) in the setting of COVID-19 illness and post-procedural decreased oral intake.

Case Report

A 74-year-old white male with a past medical history of bullous pemphigoid, heart failure, type 2 diabetes, chronic kidney disease stage 4, atrial fibrillation, and hypertension was admitted for a malodorous right first metatarsal head wound (day 0). Magnetic resonance imaging (MRI) was not suggestive of osteomyelitis and treatment with intravenous (IV) piperacillin/tazobactam plus vancomycin was initiated for a diagnosis of diabetic foot infection. Blood cultures collected prior to antibiotic initiation resulted with methicillin susceptible Staphylococcus aureus (MSSA), therefore, antibiotics were de-escalated to cefazolin 2 grams IV every 12 hours based on renal function and Infectious Diseases (ID) consult service recommendations. On day 5, antibiotics were broadened to IV vancomycin and IV cefepime due to an increased WBC count (19.6 k/µL from 13.9 k/µL on admission) and fever (100.4 °F). On day 8, the patient underwent an angioplasty revascularization procedure. Antibiotics were de-escalated back to cefazolin on day 13 per ID consult service recommendation as patient remained hemodynamically stable and medication penetration to the wound was anticipated to be improved after revascularization. On day 15, the patient underwent right great toe amputation. No cultures were obtained intraoperatively, however, proximal margins of toe resection were noted as concerning for osteomyelitis. ID planned to complete 6 weeks of cefazolin therapy and the patient was scheduled to have a right internal jugular (IJ) tunneled catheter placed for antibiotic self-administration.

On day 17, an INR obtained prior to IJ catheter placement was found to be elevated at 5.1 (with a prior baseline INR of 1.3). In response, the patient was given a single 2.5 mg dose of oral vitamin K. The INR decreased to 2.0 by day 19, after which an IJ tunneled catheter was placed. Upon discharge (day 23), the patient was continued on cefazolin 2 grams IV every 12 hours.

One day after discharge (day 24) the ID pharmacy resident was contacted by the patient’s caregiver requesting further instruction on IV antibiotic therapy and reporting inadvertent administration of double the prescribed dose of cefazolin (4 grams) for two doses. Both the patient and the caregiver were provided additional counseling on the proper administration of cefazolin. However, later that evening, the patient presented to the emergency room due to concerns of bleeding around the IJ tunneled catheter site and was readmitted to the medical center for further management. Family and social history were appropriately documented as non-contributory. Pertinent labs at readmission are reported in Table 1, most significant for: INR > 22.5, PT > 150.0, PTT 151.7 seconds, and WBC 14.0 k/µL (increased from 10.9 k/µL at prior discharge). Of note, despite being fully vaccinated, the patient tested positive for COVID-19 on day 24 (previously negative on day 0) but reported only mild symptoms and did not require or receive COVID-19-related treatments (supplemental oxygen, corticosteroids, monoclonal antibodies, antivirals, etc.).

Table 1.

Laboratory test results upon readmission (day 25).

Laboratory Test	Result	Reference Range	Units
INR	>22.5	.9 - 1.1	n/a
PT	>150.0	12.2 - 14.4	seconds
PTT	151.7	24.0 - 35.0	seconds
Hemoglobin	7.5	13.0 - 17.0	g/dL
Fibrinogen a	659	200 – 480	mg/dL
Serum creatinine	3.79	.67 - 1.17	mg/dL
WBC	14.0	4.0 - 11.0	k/µL
D-dimer	.55	</= .49	mcg/mL
Platelets	261	130 - 400	k/µL
COVID-19	Positive	Negative	n/a
Apixaban level b	218	22-177	ng/mL

^aDrawn on day 26.

^bPredicted steady state trough (lab test not cleared or approved by FDA and used only for clinical purposes).

Given available literature, prior INR responsiveness to vitamin K administration, and reported poor nutritional intake, intravenous vitamin K was administered as a one-time dose of 10 mg on day 25. The patient’s outpatient apixaban was held upon admission to decrease risk of further bleeding. An INR ordered for the day following intravenous vitamin K administration returned at 1.6. ID was consulted upon rehospitalization and recommended discontinuing cefazolin and initiating daptomycin while the patient was admitted to the hospital. Daptomycin was subsequently switched to oritavancin to complete the remaining duration of therapy. No further bleeding was noted and apixaban was resumed as outpatient the evening of discharge. An INR drawn one week after intravenous vitamin K administration (day 32) returned at 2.0, demonstrating sustained improvement in coagulopathy. Vitamin K doses and resulting INRs are provided in Figure 1.

Figure 1.

INR trend throughout cefazolin treatment.

Discussion

Cefazolin and INR

Vitamin K deficient coagulopathy related to cefazolin use has been previously summarized in several case reports (Table 2).^1-7 There are two hypothesized mechanisms for this: one theory is that this first-generation cephalosporin depletes the intestines of vitamin K producing bacteria; a second theory is that the side chain methyl-thiadiazole (MTD) in cefazolin interferes with the metabolism of vitamin K through inhibition of hepatic vitamin K epoxide reductase.^2-7 The eventual impact of either proposed mechanism would be prolonged PT/PTT, elevated INR, and increased risk of bleeding. Overall, the frequency of cefazolin-induced coagulopathy is unknown. Based on previously published case reports of coagulopathy, the effect of cefazolin on coagulation labs appears to be dose- and/or duration-dependent, with patients who have renal failure or nutritional deficiencies also at potentially higher risk.^7,8 In most cases, coagulopathy improves 1-2 days after discontinuation of cefazolin and in response to supplementation with vitamin K. However, the optimal route and dosing of vitamin K for cephalosporin-induced coagulopathy has not been studied with reported dosing ranging from 2 mg oral to 10 mg IV. Regarding the current case, INR elevations while on cefazolin occurred early in therapy but were more profound later in the treatment course and in the context of accidental, supratherapeutic cefazolin dosing. These observations, in addition to the sustained improvement in INR seen after vitamin K administration and cefazolin discontinuation, are generally concurrent with proposed mechanisms and prior case reports of cefazolin-induced coagulopathy.

Table 2.

Case report summary of cefazolin-associated INR elevation.^2-7

Case and Year	Patient age (years)/sex	Indication	Cefazolin dose	Peak INR/PT	Vitamin K Dosing	Anticoagulation
Present case	74/M	Osteomyelitis	2 g q12 h, then 4 g for 2 doses	INR >22.5, PT >150s	2.5 mg PO for initial elevation, 10 mg IV for significant elevation	Apixaban 5mg PO BID
Ngiam et al., 2021	72/M	Infective endocarditis	1 g q12 h	INR 8.29	2 mg PO x2 days	—
Shaikh et al., 2013	63/M	Surgical prophylaxis	—	INR 4.2	IV vitamin K x1 (unspecified dose)	Dalteparin 5000 units SQ daily
Chung et al., 2008	50/F	Bacteremia	1 g q24 h	INR 4.0	10 mg IV x1, then 5 mg PO x2 days	Heparin 5000 units SQ q8h
Kuypers et al., 2002	45/F	Surgical prophylaxis	2 g for 3 doses, then 6 g daily for 3 days	PT 74.6s Normotest <10%	10 mg IV x2 days	Enoxaparin 20mg SQ daily
Kurz et al., 1986	26/M	Surgical prophylaxis	2 g q12 h	Normotest 17%	Unspecified dose/route	—
Shimada et al., 1984	79/F	Pneumonia	3 g daily	PT 65.4s	10 mg IV x1	—
	87/F	Colitis	2 g daily	PT 15.0s	10 mg IV x1	—
	71/F	Pyelonephritis	2 g daily	PT 21.3s	10 mg IV x1	—
	87/F	Pneumonia, Urinary tract infection	3 g daily	PT 27.7s	10 mg IV x1	—
	74/M	Pneumonia	2 g daily	PT 28.2s	10 mg IV x1	—

Apixaban and INR

Factor Xa inhibitor DOACs can interfere with standard coagulation assays in a concentration-dependent manner. ⁹ Due to the mechanism of action, factor Xa inhibitors can prolong PT, INR, and PTT, but observed changes are typically small and variable, especially with apixaban.^9,10 A 2017 CHEST systematic review found that PT/INR and PTT were inadequately sensitive to apixaban as results may remain within normal limits even in the setting of significantly elevated apixaban serum concentrations. ¹⁰ However, prolongation of PT and/or PTT in a patient with known factor Xa inhibitor exposure should be assumed to be secondary to drug effect until proven otherwise with other laboratory assessments. ⁹ Because of the variability and lack of sensitivity of PT/INR and PTT to apixaban when administered at approved doses, manufacturer labeling recommends against the routine use of these coagulation labs for clinical monitoring. Apixaban calibrated anti-factor Xa assays provide the best correlation with drug plasma drug levels and can be used as a surrogate marker when drug monitoring is needed (though universally established therapeutic ranges are not available). ¹⁰ The current case presented with a PT > 10x ULN, INR > 18x ULN, and PTT > 4x ULN. Apixaban as a sole contributor to these abnormalities is highly unlikely in a patient compliant to the prescribed dose and in the absence of substantial drug-drug interactions. Prior laboratory assessments from past hospitalizations and collection of apixaban steady-state trough support this determination. Other contributing factors were considered as more clinically significant for this case.

COVID-19 and INR

The observed impact of COVID-19 on routine coagulation parameters in published literature is variable. While INR, fibrinogen and D-dimer may be affected, others, like aPTT, often remain normal regardless of illness severity. ¹¹ Of note, D-dimer can also be markedly elevated in a variety of other clinical circumstances. As such, despite the anticipated pro-coagulopathy of COVID-19, the prognostic value of these biomarkers has been explored by and incorporated in therapeutic resources with limited consensus.^12,13 Furthermore, differentiation of COVID-19 coagulopathy from other conditions, such as DIC, can be challenging. In principle, early COVID-19 infection is more often characterized by normal platelet values and increased levels of fibrinogen rather than the decreases in both seen with DIC.^13-15 However, more severe COVID-19 infections may bear more similarities to DIC with a corresponding worsening of patient outcomes.^13-15 Taken altogether, our patient’s laboratory panel and overall mild illness point to contributing factors outside of COVID-19 alone. While the D-dimer is significantly elevated, this lab may be impacted by the patient’s recent operation, current infection, and renal impairment. The high fibrinogen and normal platelet count also do not support DIC as a primary condition. And while alterations in INR and PTT are possible in COVID-19 illness, the substantial elevations seen in this case do not reflect the marginal to absent changes cited in literature as typical of early or mild illness, making COVID-19 an unlikely sole cause of the observed laboratory abnormalities.

Malnutrition and INR

Blood coagulation is controlled by the balance between procoagulant and anticoagulant systems. Several nutrients affect coagulation, including vitamin K which is necessary for the synthesis of most clotting factors with the exception of factors III, IV, and VIII. ¹⁶ Vitamin K- dependent clotting factors are divided into anticoagulant (protein C and protein S) and procoagulant factors (II, VII, IX, and X). ¹⁷ Vitamin K procoagulant factors undergo vitamin K mediated post-translational modification for conversion into functional proteins. Deficiency of vitamin K from chronic malnourishment can affect all three coagulation pathways, potentially resulting in prolonged PT, INR, and PTT even in the absence of pharmacological anticoagulation. Furthermore, though the bioavailability of apixaban is not principally affected by food, apixaban is approximately 87% bound to plasma proteins, primarily albumin. ¹⁸ Albumin concentrations can be affected by various conditions that affect production or clearance of albumin, including heart failure, cirrhosis, nephrotic syndrome, and malnutrition. Chronic nutritional deficiencies affect albumin synthesis through a decreased supply of amino acids and trace minerals. ¹⁹ Although hypoalbuminemia caused by chronic malnutrition has not been directly evaluated in patients on DOACs, theoretically, low albumin levels may result in increased drug exposure and in turn affect coagulation assay results. Nutritional deficiencies were not overly apparent in our patient and despite poor oral intake being noted, albumin levels were normal (vitamin K level was not measured). Despite normalization of PT and INR after administration of IV vitamin K, vitamin K deficiency secondary to malnutrition was likely not an independent contributor to this patient’s elevated PT and INR.

Conclusion

Disruptions in mechanisms of coagulation can result from a variety of contributing factors. While literature review can further define possible causes, an individualized evaluation is key to managing co-occurring conditions and therapies. In this instance, early detection of patient reported PICC line bleeding while on cefazolin with assessment of medication adherence and laboratory parameters were essential to care plan development. COVID-19, concurrent apixaban, and malnutrition were all evaluated as potential reasons for observed INR elevations; available literature and patient-specific qualifiers (mild viral illness, prior INR stability on DOAC therapy, and absence of apparent severe nutritional deficiencies) established each as an unlikely primary etiology. As seen in prior case reports and related research, the significant INR prolongation was definitively managed with vitamin K administration and conversion of antimicrobial therapy to an alternative agent, highlighting cefazolin as the most likely contributing factor. For the clinician, it is worth noting that incorporating patient interview questions to detect PICC line bleeding with thoughtful consideration of planned INR monitoring may be important components of extended intravenous cefazolin therapy.

Key Points

• Cefazolin may cause vitamin K deficient coagulopathy resulting in prolonged INR, PT and PTT.

• Multiple medical comorbidities can independently contribute to hypercoagulability and clotting factor consumption, making determination of distinct etiology difficult.

• Vitamin K may be considered to correct an elevated INR secondary to cefazolin.