Shock is a life-threatening condition characterized by inadequate global oxygen delivery relative to demand [1]. It is a common problem in hospitalized patients, can result from various etiologies, most frequently septic shock, and is a leading cause of mortality worldwide [2]. The infusion of vasopressors is a critical component of shock management, with recommendations not to delay use of vasopressor administration until fluid resuscitation is completed but to initiate vasopressors early to achieve target mean arterial pressure [3]. While international guidelines recommend rapid initial administration of these medications through peripheral intravenous (PIV) catheters for a short period if central venous catheters (CVCs) are unavailable, they recommend eventual transition to CVCs for continued administration due to concerns about extravasation and local tissue injury with PIV administration [4].

However, CVCs are often unavailable or challenging to use in many resource-variable settings due to supply chain issues, limited clinician training, and infection control concerns [5]. Recognizing these challenges, several studies have been conducted to assess the safety of vasopressor infusion through PIV catheters as an alternative to CVCs by evaluating the frequency of adverse events of peripherally administered vasopressors [6,7]. A 2019 retrospective single-center cohort study by Lewis et al [8] evaluated 202 patients who received peripheral vasopressors, primarily norepinephrine, with a median infusion duration of 7.5 hours. The incidence of extravasation events was 4%, all managed conservatively. An Australian cohort study of 202 patients reported a 23.2% complication rate with peripheral vasopressor administration, mostly involving phenylephrine infusions for a median duration of 10 hours [9]. In a cancer hospital in India with resources similar to those of the proposed study setting, 122 adults received peripheral vasopressors for a median duration of 11 hours, with only 1 case of extravasation-related tissue injury [10]. A 2022 prospective study in a Pakistani pediatric intensive care unit (ICU) reported a 2.2% extravasation injury rate among 369 patients receiving peripheral vasopressors, primarily epinephrine, for a median duration of 24 hours [11]. A recent prospective cohort study conducted across multiple hospitals in Addis Ababa, Ethiopia, enrolling adults with circulatory shock also found low rates of extravasation injury (3 in 250 patients who received peripheral vasopressors). None of the extravasation injuries occurred in patients who received peripheral vasopressors for less than 5 days [12].

When available, CVCs are not without complications [13]. A study from a high-volume facility in South Africa found a complication rate of 18% among more than 1000 patients receiving a CVC [14]. A recent meta-analysis of 130 studies found a 3% rate of major complications associated with CVC placement—including arterial cannulation or puncture, pneumothorax, infection, and deep vein thrombosis [15]. In the pediatric population, as many as 1 in 4 CVCs fail before completion of therapy [16]. Placement of a CVC is the main risk factor for development of deep vein thrombosis in children, with one study finding an incidence of CVC-related thrombosis of 30.3% [17,18].

Even in well-resourced settings in which CVCs are readily available, there may still be a role for peripheral vasopressor administration. Studies have shown that early initiation of vasopressors in patients with septic shock is associated with increased survival [19]. It generally takes more time to place CVCs as compared to PIV catheters, and depending on the location of the CVC, radiographic confirmation of placement must be obtained before it can be used, further delaying initiation of vasopressors. Thus, multiple previous studies have been conducted to describe and assess the prevalence of complications of peripheral vasopressor administration. A 2020 systematic review and meta-analysis by Tran et al [20], which included 8 observational studies and 1 randomized controlled study with a total of 1835 adult patients who received peripheral vasopressor administration, reported 7% adverse event occurrence, with only 4% of those being major events. Similarly, another systematic review and meta-analysis by Owen et al [21] in 2021, which included both adults and children, reported a 1.8% complication rate for PIV catheters. A 2015 systematic review by Loubani and Green [22], which included primarily case reports, described published reports on local tissue injury or extravasation events. The average duration of peripheral vasopressor administration was 55.9 hours before local tissue injury and 35.2 hours before extravasation [22]. Most recently, a secondary analysis of the Crystalloid Liberal or Vasopressors Early Resuscitation in Sepsis trial evaluated the route of vasopressor initiation and continuation of peripheral vasopressors beyond 6 hours and found that most patients had vasopressors initiated peripherally and continued past 6 hours, with rare low-grade complications [23,24]. Importantly, none of these reports were from the African continent.

Collectively, these studies suggest that short-term peripheral vasopressor administration may be a reasonable alternative when CVCs are unavailable. However, most studies have evaluated relatively limited patient populations and short infusion times. Additionally, most studies have been retrospective, limiting the ability to capture administration and complication details comprehensively. To date, few studies have thoroughly evaluated the safety profile of prolonged peripheral vasopressor infusions regarding duration or concentration, particularly in resource-variable settings in which PIV catheter use is common. Finally, to our knowledge, no study has directly compared the safety of vasopressor administration via PIV catheter vs CVC.

At AIC Kijabe Hospital (KH) in Kenya, it is standard practice to use PIV catheters for the duration of a patient’s vasopressor requirement, often lasting several days, regardless of the agent used or concentration and dose required. There is no maximum concentration or dose for which it is required or standard to switch to a CVC. This is because CVCs are infrequently used due to cost, supply chain issues, and limited clinician training. Therefore, this study aims to address knowledge gaps by documenting the complication rates among patients receiving vasopressor infusion through PIV catheters.

Consecutive enrollment began in October 2023 and is currently ongoing. We anticipate that the time to enroll the number of patients required to reach our power goal will be 24 months. Once this goal has been reached, we estimate that reviewing, analyzing, and filling in the missing data will take 6 additional months. During this time, there will be periodic review of the data collected to ensure accuracy, as well as periodic analysis to assess for the occurrence of our primary outcome.

This is a prospective, investigator-initiated, single-center observational cohort study that plans to enroll and follow up on patients receiving vasopressor infusions via PIV catheters and CVCs to describe and compare the incidence of complications. Our hypothesis is that there will be few severe complications of administration of vasopressors via PIV catheter in this setting and that the safety of vasopressor administration via PIV catheter will be noninferior to that of administration via CVC.

This study is being conducted at a rural Kenyan hospital in a resource-variable setting. KH was chosen as the study site for a variety of reasons. Most importantly, the current usual practice at KH is the use of PIV catheters for administration of vasopressors regardless of agent, duration, or rate used. CVCs are not commonly placed primarily for vasopressor administration but, rather, are more commonly placed for other indications, such as total parenteral nutrition, hemodialysis, or chemotherapy, and are sometimes subsequently used for vasopressors if available [25]. This is in contrast with current standard practice in many high-resource settings, where vasopressor administration is frequently a primary indication for CVC placement. In addition, the duration of vasopressor administration via PIV catheter at KH can be multiple days. This differs from current international guidelines that recommend insertion of a CVC to continue vasopressor administration for prolonged infusion times (>6 hours) [4]. Thus, KH provides a unique setting for this study to help provide evidence on the adequacy and safety of peripheral vasopressor administration in resource-variable settings in which CVC placement may not be feasible.

Compared to previous studies, this is the first study to include both adult and pediatric patients, resulting in a relatively large sample size. In addition, patients are enrolled not just from the ICU but also from the HDU, and they include a mixed ICU population with broad etiologies of shock not limited to septic shock, which improves the generalizability of the results. As mentioned previously, many patients at KH have a prolonged duration of vasopressor infusion via PIV catheter. This contrasts with previous studies that have mostly looked at short infusion times, which not only provides an ideal setting for this study but also reflects the reality of resource-variable setting practice. Finally, this is a prospective observational study with enrollment ongoing in real time, allowing for more accurate and comprehensive data collection on vasopressor administration and potential complications as they arise.

There are several limitations to this study, the most important of which is that, as this is a prospective observational study without randomization, there will be residual confounding variables and differences in baseline between the PIV and CVC groups. As patient populations receiving CVCs generally have higher severity of illness and mortality than those without CVCs in critical care studies, we will not compare the PIV and CVC groups based on these clinical outcomes. In addition, as there is no blinding, a heightened awareness of potential complications may result in a lower threshold to detect a complication and change the IV site, thereby overestimating the complication rate. We have categorized the complications as mild, moderate, and severe to better account for this. Finally, this study is being conducted in a resource-limited hospital with a limited electronic medical record, with potential for limited or missing data.

This study aims to address the critical knowledge gap in understanding vasopressor administration safety in resource-variable settings in which central venous access may be limited or unavailable. By studying vasopressor delivery in a setting in which PIV catheter use is the primary method of delivery rather than the exception, we hope to provide insights into alternative approaches to critical care medication delivery in resource-variable settings.