The Art of Medicine: Balancing Evidence with Clinical Judgment

Clinical Scenario:

Ms. S. is a 76-year-old female with a history of DM, CAD s/p stent, CKD III, and severe depression who is bought in by her family for severe uncontrolled depression to the point where she had become inactive, essentially bed-bound for one week and lost 25 lbs over two months. The original plan was to admit her to Geri-Psych to manage her depression, however laboratory evaluation in ER revealed mild rhabdomyolisis and acute on chronic kidney insufficiency (Creat 2.1). Thus, the patient was admitted to the medical service for further evaluation and management. While on the medical service she was kept on all her home medications, which included ASA 81mg and Plavix 75mg. In addition, she was started on standard venous thromboembolism (VTE) prophylaxis therapy with low dose unfractionated heparin (LDUH or UFH) 5000U TID as well as intermittent pneumonatic compressions (IPCs or SCDs). Ms. S. was also being seen by a physical therapist and nutritionist while in the hospital for her deconditioning and malnutrition, respectively. By Hospital Day 2 (HD2), after fluid resuscitation, the patient’s renal function improved to baseline (Creat 1.6, CrCl 35), and she was medically cleared to be transferred to inpatient psych. On HD3, there was a bed available and arrangements were made for the transfer. A few hours prior to transfer, the patient was assisted out of bed to a chair for her lunch. Shortly after, she started complaining of right axillary discomfort. The exam this time was unremarkable and she was given a heating pad and some Tylenol. However a few hours later, it was noted by the nursing staff that the patient’s right breast was twice a large as the left one and she had swelling and brusing around the axillary area. Imaging showed an expanding hematoma as a result of a muscle tear and labs revealed a ten point drop in hematocrit.

There is no doubt that prevention of VTE by use of thromboprophylaxis (medical or mechanical) has clear benefits in both acute and long-term morbidity and mortality, especially in terms of symptomatic DVT/PEs, fatal PEs, length of stay, cost, post-thrombotic syndrome, etc. In fact, the ACCP (American College of Chest Physicians) recently released their most updated guidelines on this topic which details their evidence-based recommendations as it applies to many different medical/surgical scenarios. While the evidence emphasizing the use of primary thromboprophylaxis is strong and supported by several randomized clinical trials (RCT), it does not specify what frequency or doses of these medications (LDUH, LMWH or enoxaparin, fondaparinux, etc) are preferred in terms of efficacy and safety.

I wanted to bring up this topic because a significant number of inpatients admitted to the medical service are frail, elderly patients with some level of renal dysfunction. These are also the same patients who are at high risk for developing clinically significant VTE. I don't want to focus on which patients are at low vs. moderate vs. high risk for VTE and should receive thromboprophylaxis; this is an entirely different topic which is again outlined in the ACCP guidelines. Instead, I wanted to address some concerns that come to mind every time I encounter a patient that is either elderly, has renal dysfunction, has a low BMI, or is on ASA/plavix, etc, who is started on thromboprophylaxis with LDUH 5000U TID or LMWH 40mg qd. While these are the standard doses/frequency that are acceptable for most of the inpatient population, using such doses/frequency can greatly increase the bleeding complications in the high-risk population mentioned above. After encountering a couple of patients with such complications, I asked myself, "Where is the data that states that every patient who needs VTE prophylaxis must receive it at this standard dosing (LDUH 5000U TID or LMWH 40mg qd) for it to be effective?"

In my search, I was not able to find a RCT that directly compared LDUH TID vs. BID or LMWH 40mg vs. 30mg in terms of efficacy compared to safety. However, I did come across a metaanalysis that reviewed 12 existing trials including approximately 8,000 patients in an effort to compare BID vs. TID heparin, specifically in the general medical population. Without getting into the nitty-gritty aspects of this study, let's just say the review had many limitations. I did, however, like the overall message that is suggested by the conclusion: "BID heparin dosing causes fewer major bleeding episodes, while TID dosing appears to offer somewhat better efficacy in preventing clinically relevant VTE events. Practitioners should use underlying risk for VTE and bleeding to individualize pharmacologic prevention. "

What does this mean? It means that while clinical trials enable us to practice "evidence based medicine,” we cannot overlook upon whom this evidence is based. The literature supports the routine use of thromboprophylaxis (medical or mechanical) in order to prevent VTE, and every patient should be evaluated as such. However, our decision on what dosing and frequency should be individualized for every patient based on their age, renal function, weight, co-morbidities, and medications, all of which contribute to their overall risk to benefit ratio (VTE vs. bleeding complications). For example, while an 80-year-old patient with decompensated heart failure or severe respiratory illness, who is mostly bed bound, definitely needs VTE prophylaxis, you may consider using either LDUH 5000 U BID (low BMI) or LMWH 30mg qd (renal dysfunction) in addition to mechanical methods (IPC/SCDs). This may be efficacious in VTE prevention while significantly decreasing the risk of bleeding complications (see section 1.4.6 of ACCP guidelines).

Luckily the patient in the above scenario did well. However, she did end up requiring 3 units of packed red blood cells and her hospital stay was prolonged by three days, not to mention the time it will take for the hematoma to resolve. Such situations certainly remind me to individualize evidence based guidelines to each particular patient.