Sheth KN, Smith EE, Grau-Sepulveda MV, Kleindorfer D, Fonarow GC, and Lee H. Schwamm LH. Drip and Ship Thrombolytic Therapy for Acute Ischemic Stroke: Use, Temporal Trends, and Outcomes. Stroke. 2015
Finding ways to improve access to stroke care is of high priority. The next promising stroke treatment is rendered useless unless accessible to stroke patients. In light of this, drip and ship has been one of the many methods to increase IV-tPA access to the general population. Smaller hospitals that do not feel comfortable managing patients will often transfer to larger centers after initiating IV-tPA. Additionally, there is often transfer with the intention of potentially bridging with IA therapy. With the advent of the recent MR CLEAN results, we will definitely see this trend increase. While many places have already become knee deep in the drip and ship method, the questions that still loom at large include the following: 1) Is it a safe process? 2) Are there differences in the populations? 3) Are outcomes affected?
Sheth et al evaluate these questions in a retrospective fashion using GWTG data from April 2003 to Oct 2010. They included patients who received IV-tPA under 3 hours. Excluded were sites with fewer than 30 IV-tPA patients, inpatient strokes, patients receiving experimental therapies. Ultimately, 44,667 patients were treated with IV-tPA. 10,475 (23.5%) were by way of drip and ship. Baseline characteristics were unbalanced with regards to stroke severity (first NIHSS median 12.9 for front-door patients and 11.0 for drip& ship [significant]), arrival between the hours of 7 am to 5 pm (47.2% vs. 28.7% significant), age 70.1 vs 67.3 years (significant), and sex (male 49.9% vs. 46.3% significant). With regards to outcomes adjusted for NIHSS, hospital mortality, sICH, independent ambulation and discharge home all favored front door patients; significant.
The flaw of this study lies in the fact that the baseline characteristics of the front door vs. drip and ship group are imbalanced in so many different categories. Statistically, Sheth et al attempt to adjust for these differences in their model. Due to the large number of subjects evaluated in the study, there is ultimately some credence to the results. It tells us that as a whole, drip and ship patients may do slightly poorer than their doorstep counterparts. However, their baseline NIHSS is lower than their front door counterparts, which although initially baffling, is explained by the fact that scales documented for drip & ship patients typically occur upon arrival to the second hospital. Thus the drip & ship patients have received the IV-tPA for some time already, and early responders may confound this comparison of baseline NIHSS.
There are many other instances that we can point out in terms of weaknesses due to the inherent nature of being a retrospective study. Nonetheless, the study still makes a powerful observation that drip and ship method for IV-tPA treatment is a definite reality. 25% of IV-tPA’s were given in a drip and ship manner. Although the data does appear to show that drip &ship patients do slightly worse with the above outcomes, compared to their front door counterparts, the retrospective nature of this study and baseline data capturing likely are confounding the results. I definitely agree with the encouragement of this paper to push for improvement of data capturing by targeting quality improvement projects and performance measures. After all, even if outcomes are not as good as front door patients, drip and ship still has the potential for better outcomes than no intervention at all.