Construction of a nursing management program for early fluid resuscitation in patients with acute pancreatitis: a Delphi study in China

The PubMed, Medline, Web of Science, Cochrane Library, CNKI, and Wanfang databases were systematically searched from the inception of the databases up to June 2023. The search strategy involved a combination of subject words and free words. The methods employed included searching for relevant systematic evaluations, meta-analyses, and pertinent original research papers. Titles, abstracts, keywords, and references of these papers were analyzed to identify keywords for the literature search. Synonyms were also expanded based on the retrieved articles. The primary search terms used were “acute pancreatitis,” “fluid resuscit*,” “fluid therap*,” “intravenous fluid*,” “intravenous resuscit*,” “nursing,” “Delphi method,” “indicators,” “programme*,” and others. Studies related to acute pancreatitis published in either English or Chinese were considered for inclusion. Additionally, we incorporated relevant sources such as the Expert Consensus on Prevention and Interruption of Emergency Care in Severe Acute Pancreatitis [14], the Chinese Guidelines for the Diagnosis and Treatment of Acute Pancreatitis (2021) [1], the Expert Consensus on Emergency Diagnosis and Treatment of Acute Pancreatitis [15], the 2019 World Society of Emergency Surgery (WSES) guidelines, and more [16‐19]. Following the SMART screening principle in health management performance evaluation theory [20], relevant program indicators were initially extracted. The entire literature search and screening process was done independently by J.L and S.J.M, who wrote the meta-analysis, respectively, and a third party was asked to adjudicate on discrepancies, resulting in a pool of indicator entries for the nursing management program for early fluid resuscitation in acute pancreatitis.

A purposive sampling approach was utilized, and sample information saturation was the guiding principle for semi-structured interviews conducted with eight specialists and five specialist nurses from related fields in two comprehensive tertiary hospitals located in two provincial capital cities in China. The interview outline was established through multiple discussions within the group and included the following key questions: (1) What components do you believe should be incorporated into the workflow and specific content of early fluid resuscitation nursing care for acute pancreatitis? (2) In what ways do you propose the quality of early fluid resuscitation care for acute pancreatitis should be assessed, and what criteria should be used? (3) Which nursing factors do you anticipate will influence the quality of early fluid resuscitation in patients with acute pancreatitis? (4) What shortcomings do you perceive in the current approach or content of early fluid resuscitation for acute pancreatitis? (5) What actions do you recommend to enhance and improve the quality of fluid resuscitation care for acute pancreatitis? The sample size for the interviews was determined by the criterion of adding two additional participants after reaching a point where repeated information was obtained, and no new themes or insights emerged during the interviews. By combining the previously collected terms and indicators, which were identified through on-site research, and subsequently categorizing, refining, and summarizing them, a preliminary draft of the early fluid resuscitation nursing management program for acute pancreatitis was developed. This draft included 4 primary indicators, 12 secondary indicators, and 37 tertiary indicators. To ensure its quality, four experts who possessed extensive experience in the diagnosis, treatment, and nursing care of acute pancreatitis evaluated the readability and feasibility of this preliminary program draft.

The experts who took part in the Delphi survey were drawn from diverse regions and organizations across China. The criteria for their inclusion were as follows: Education: Possession of a bachelor’s degree or higher. Title: Holding a position of associate senior or higher. Work Experience and Years of Experience: A minimum of 10 years of experience in clinical or nursing work within the relevant specialty area, specifically in a tertiary general hospital. Voluntary Participation: Willingness to participate in the survey on a voluntary basis. Fifteen experts were ultimately included, 10 (67%) of whom were engaged in clinical medical work and 5 (33%) of whom were engaged in clinical nursing work. The experts came from different regional hospitals such as Tongji Hospital of Tongji Medical College of Central China Medical University, Xuanwu Hospital of Capital Medical University, Gulou Hospital of Nanjing Medical University, Zhujiang Hospital of Nanfang Medical University, Baiqiu’en Hospital of Shanxi Province, the First Hospital of Shanxi Medical University and Yuncheng City Central Hospital. The average age was 44 ± 4 years, and the average work experience was 16 ± 4 years. The demographic data of the individuals included in the Delphi panel are shown in Table 1.

Based on the elements of the program obtained in the first draft, a questionnaire was developed for experts on early fluid resuscitation care programs for acute pancreatitis. The questionnaire consisted of three parts: general information about the expert, the main text, and a self-evaluation form for the expert. The questionnaire used a 5-point Likert scale ranging from 5 (very important) to 1 (very unimportant) to assess the degree of importance of each indicator. Columns were included for “modification” and “proposal for adding new program indicators” to allow experts to provide input on modifying or removing existing indicators and suggesting new indicators not originally included in the questionnaire. Experts had the opportunity to propose modifications, deletions, or additions to the indicators. Two rounds of Delphi expert correspondence were carried out in July and August 2023. Both rounds were distributed offline or via email by the subject leader, and participants were requested to return the completed questionnaires either in paper format or through email within 1 week. The Delphi method was employed to calculate the weights of the indicators [21] Indicators with a mean importance rating of ≥ 4.0 and a coefficient of variation of ≤ 0.25 were retained as screening criteria. Additionally, there needed to be at least 75% agreement among the experts. In cases where an indicator met the criteria for deletion, the group members discussed it before making a final decision to prevent the removal of important indicators. We added, merged and modified some of the entries according to the experts’ opinions and reformulated the questionnaire for the second round of correspondence. The two rounds of correspondence questionnaires were collated and analyzed at the end of the two rounds to ultimately form the nursing management plan for early fluid resuscitation in acute pancreatitis patients.

Statistical analysis employed SPSS 25.0 software. Descriptive analyses utilized means, standard deviations, coefficients of variation, and proportions. The effective recovery rate of the questionnaire conveyed the degree of expert positivity. Expert authority was assessed based on judgment and familiarity with the issue. The Kendall harmony coefficient quantified the degree of coordination of expert opinions [22].

The study received approval from the hospital’s medical ethics committee (ethics approval number YXLL-2023–113). Participants were provided with assurances regarding the voluntary nature of their participation and the anonymity and confidentiality of their data.

In this study, 15 and 14 questionnaires were distributed during the two rounds of expert consultation, and 15 and 14 valid questionnaires were collected, respectively, for a valid recovery rate of 100%. Thirteen experts—87% —provided specific comments in the first round of correspondence, and three experts—21% —provided comments in the second round of correspondence.

The degree of expert authority (Cr) is determined by the basis for judgment (Ca) and the expert’s familiarity with the program indicators under investigation (Cs). The formula for calculating Cr is as follows: Cr = (Ca + Cs)/2. In this study, the experts’ basis for judgment (Ca) in the two rounds of correspondence was 0.993 and 0.993. The familiarity with the program indicators (Cs) was 0.947 and 0.957. Therefore, the coefficients of authority (Cr) were 0.970 and 0.975 for the respective rounds.

It is expressed through the coefficient of variation (CV) and coordination coefficient. The CV of the two rounds of expert correspondence in this study ranged from 0.05 to 0.21 and 0.00–0.20. The coordination coefficient is evaluated by Kendall’s harmony coefficient, and the expert coordination coefficients of all levels of indices in this study are 0.166–0.335 and 0.189–0.364, respectively. The P values of the first, second, and third level indices are < 0.05 according to the test of Kendall’s harmony coefficient, which is statistically significant. The degree of coordination of the expert correspondence is considered good (Table 2).

In the initial round of the Delphi survey, experts evaluated the initial draft of the nursing management program for early fluid resuscitation in acute pancreatitis patients. This evaluation included 4 primary indicators, 12 secondary indicators, and 37 tertiary indicators, as detailed in Table 3. The first round garnered fifty-one comments. Subsequently, the program underwent revision and refinement by the research team. This process involved a thorough examination of expert recommendations and ensuing discussions. During this revision, three new protocol indicators, specifically those pertaining to the use of anticoagulants, use of antihypertensives, and the rate of blood glucose decline, were not endorsed, as they fell beyond the study’s scope. Additionally, one indicator was divided, two indicators were combined, one new indicator was introduced, and eleven adjustments were made to the wording. Furthermore, two items that did not meet the 75% protocol threshold were removed. All the program indicators that were retained emerged from the second round of expert consultations.

During the subsequent round of correspondence, consensus was achieved for all program indicators, guided by predefined criteria. Only minor wording adjustments were made, and no additions or deletions occurred. Consequently, a total of five primary indicators, 11 secondary indicators, and 36 tertiary indicators were conclusively established, as outlined in Table 4.

The program applies to emergency, ward, and intensive care units. In the initial round of communication, experts recommended changing “time of diagnosis” to “time of onset.” They believed that “time of onset” would better assist medical staff in assessing the patient’s condition and progress initially. It also guides efforts to enhance disease literacy and awareness in the population before or after hospitalization, minimizing the time gap between illness onset and medical treatment. This approach also helps direct the development of initiatives to enhance disease literacy and awareness in the prehospital or posthospital population, further reducing the time gap between illness onset and when patients seek medical treatment. “Medical history” includes the history of the presenting illness and past medical history, aiding medical staff in identifying potential causes of AP and contraindications to rehydration. Rehydration protocols often consider body weight [30]. Even bedridden patients should have recent weight information obtained from patients or family members, as recommended by experts, to enhance practicality. Assessing prior treatments involves determining whether patients received relevant treatment before evaluation, influencing our protocol to avoid over- or under-rehydration. Furthermore, nurses must assess AP severity in patients. According to the revised Atlanta classification [31], acute pancreatitis can be classified as mild acute pancreatitis (MAP), moderately severe acute pancreatitis (MSAP), and severe acute pancreatitis (SAP) according to severity. MAP is the absence of organ failure, local or systemic complications. MSAP is defined as the presence of transient organ failure, local complications, or worsening of comorbidities. SAP was defined as persistent organ failure, that is, organ failure > 48 h. As early-stage AP severity assessment, especially within 24 h, remains unsatisfactorily resolved, expert opinion suggested omitting the phrase “use the revised Atlanta Classification”. According to the Expert Consensus on Prevention and Intervention of Severe Acute Pancreatitis in Emergency Care developed by the China Association for the Promotion of International Exchanges in Healthcare, it is recommended that the diagnosis of “suspicious severe acute pancreatitis (SSAP)” be included in the current classification criteria in order to provide timely support and treatment [32]. The APACHE II score, Marshall’s score, Ranson’s score, and SOFA score can assist in early or retrospective AP severity assessments. Although the APACHE II score, Marshall score, Ranson score, SOFA score, etc., can determine AP severity early or retrospectively, they are not widely used in clinical practice due to complexity and inconvenience. However, the bedside index for severity in acute pancreatitis (BISAP) is considered a practical method for predicting SAP due to its simplicity and accessibility [33, 34]. Of course, these tools should not replace clinical judgment.

In fact, the first few hours after the onset of the disease are considered crucial for preventing SIRS, progression of MODS and/or worsening of pancreatic necrosis [35]. Moreover, AP is dynamic and can progress to a serious illness, which can be influenced by timely intervention [36]. Early fluid resuscitation is key for achieving the most effective goal-directed fluid resuscitation in AP patients. There were two time intervals between the onset of disease, the time to medical care, and the time to resuscitation initiation. The first time difference is not easy to control, but we can minimize the second time difference between onset and fluid resuscitation, and nurses can minimize the damage to the pancreas and systemic microcirculation induced by the inflammatory cascade response by performing fluid resuscitation immediately [37, 38]. However, nurses should clearly define the different phases of fluid resuscitation and the goals of resuscitation, and observe and initiate appropriate therapy. If the patient is transferred to the hospital 3 days after the onset of the disease, and if a more ideal fluid resuscitation has been performed before transfer, the patient should have entered a more stable stage after 3 days, and the main goal is to perform a conservative fluid balance to maintain fluid balance or negative balance, and the main risk of this stage is fluid overload; however, if the patient’s fluid resuscitation was not ideal before transfer, the window of opportunity for maximising fluid resuscitation may have passed, and it is still primarily important to make judgements about the body’s needs based on the current patient’s vital signs and relevant laboratory indicators, etc., because the fluid resuscitation process is dynamic. And communication between the nurse and the physician overseeing the fluid resuscitation results is very, very important during these processes. Given its importance, we followed the experts’ advice to divide the primary indicator “admission assessment and management” into two separate indicators: “initial in-hospital assessment” and “initial in-hospital management.” The implementation of “immediate intravenous infusion” can be enhanced through measures like medical and nursing agreement prescriptions, as well as comprehensive packages of medications and medical devices.

What fluid is chosen to establish resuscitation access, crystalloid or colloid? Most guidelines recommend a balanced salt solution such as Ringer’s lactate, which has a more favorable anti-inflammatory effect than 0.9% NaCl for SIRS and C-reactive protein control within 24 h [39]. A Stepped Wedge Cluster Randomised Trial by Dr Windsor’s team in New Zealand also demonstrated multiple clinical benefits of preferential use of balanced solutions in predicted severe acute pancreatitis [40]. Therefore, Ringer’s lactate was directly adopted in the programme, which also facilitates the stock of self-contained drugs in the nursing units of each hospital. For hydroxyethyl, a common colloid in the clinic, large randomized controlled trials have shown adverse effects such as renal impairment; thus, the current evidence does not support the use of artificial colloids such as hydroxyethyl in SAP patients [41]. The use of colloids in critically ill patients is controversial and needs to be determined by a doctor according to the patient’s situation; therefore, according to expert opinion, these patients were excluded. The rate of fluid resuscitation should be goal-oriented according to the guidelines, but there are substantial differences among the studies. The Chinese 2021 guidelines [1] specify an initial rate of 5–10 mL/kg/h for the first 24 h. However, there is no specific recommendation regarding the duration of this rate. Some sources suggest a duration of 24 h. It is important to note that administering too much fluid over 24 h has been linked to a worse prognosis [42]. 2022 The New England Journal of Medicine published an open multicenter randomized controlled trial. This trial investigated the safety and efficacy of weight-based active fluid resuscitation versus moderate fluid resuscitation in the treatment of AP. Active fluid resuscitation involved a push of 20 mL/kg for more than 2 h, followed by 3 mL/kg/h. Moderate fluid resuscitation consisted of a push of 10 mL/kg only when hypovolemia was present, followed by 1.5 mL/kg/h. The study revealed that active fluid resuscitation was associated with a higher risk of volume overload and did not lead to improved clinical outcomes [30]. Initially, the research team favored the moderate fluid resuscitation approach from this study. However, recognizing the potential risk of inadequate resuscitation when using fixed infusion rates, a more tailored approach became necessary. Volume expansion should be adjusted within the first few hours of admission based on a careful assessment of the patient’s volume status, especially in severe cases. Consequently, fluids should be initiated at a rate of 5–10 mL/kg/h. If resuscitation goals are achieved at any point during the first 24 h, the fluid rate should be reduced to 2–3 mL/kg/h [1, 43]. It is important to highlight that patients who do not exhibit a rapid clinical response within the initial 6–12 h of fluid therapy may not benefit from large volumes of fluid administration [44]. Five experts believe that reducing the fluid volume after 12 h may cause the total volume of fluid received by patients to increase, so it was discussed that patients who do not show a rapid clinical response after 6 h are advised to slow the drip rate and ask for a relevant consultation. There is also no definitive conclusion about the amount of fluid needed to replenish AP patients, but nurses should be aware that, in most cases, 2.5 4 L of fluid in the first 24 h will achieve the resuscitation goal, but there are people who may need up to 5 L or more per day in the initial phase [45].

In conclusion, acute pancreatitis represents a multifaceted and evolving disease process, highlighting the importance of personalized fluid therapy [46]. The goals and methods of fluid therapy vary depending on the stage of resuscitation, and in 2022 Dr Windsor’s team studied the safety and feasibility of fluid therapy regimens in the second and optimised phases of resuscitation in patients with predicted severe acute pancreatitis [47]. Excessive intravenous fluid administration can result in water and salt overload, potentially exacerbating the condition [48]. In the context of critically ill patients, vigilance and meticulous patient monitoring are imperative due to the potential risks associated with improper fluid management [49]. Close clinical and hemodynamic monitoring and a clear definition of resuscitation goals are fundamental. The resuscitation process requires close collaboration with the physician, and the nurse, as the sentinel and implementer of fluid resuscitation, needs to assess the patient frequently to achieve the recognised goals of expelling adequate urine output, reversing tachycardia and hypotension, improving laboratory markers, and adapting the management to the clinical findings and trends. Certain laboratory values, such as hematocrit and blood urea nitrogen, have traditionally served as markers for hypovolemia and can offer valuable insights into assessing fluid status. Elevated values upon admission and their subsequent increase during the initial 24 to 48 h may indicate inadequate fluid resuscitation [50]. However, whether rapid fluid therapy should be administered to hemoconcentrated AP patients is controversial, and studies have shown that early rapid fluid therapy is associated with an increased risk of non-invasive positive pressure ventilation in hemoconcentrated patients with severe acute pancreatitis [51]. In view of the limitations of conventional haemodynamic monitoring techniques, such as central venous pressure being an indicator of pressure affected by intra-abdominal pressure, etc., we can also use non-invasive techniques such as the use of cardiac ultrasonography to assess volume replenishment in patients with SAP. In cases of SAP, precise fluid therapy adjustments are essential. Similar to other situations requiring substantial fluid management typically encountered in the ICU, a combination of noninvasive clinical assessments, invasive hemodynamic parameters, and laboratory indicators should guide healthcare professionals and nurses during the early stages of SAP. This approach ensures that organ perfusion requirements are met through appropriate fluid administration while mitigating the adverse effects of fluid overload. Furthermore, it is crucial to note that in the specific context of SAP, complications like bowel wall edema and retroperitoneal edema can lead to the development of abdominal compartment syndrome, posing a new set of challenges [52]. Intra-abdominal pressure measurement is advocated for patients with increasing abdominal distention to monitor the possible development of septal compartment syndrome, and the test can be performed using the optimised intravesical pressure measurement technique advocated by Dr Windsor’s team [53, 54].

Nurses play a crucial role in developing and implementing health education. Effective health education enables AP patients to gain a comprehensive understanding of the information and purpose behind fluid resuscitation, enhancing their treatment adherence, particularly during the initial active rehydration phase. Nurses should communicate to patients the importance of good compliance for effective early fluid resuscitation treatment. They should refrain from making arbitrary adjustments to the drip rate to prevent any decrease in compliance, which could reduce therapeutic effectiveness. Furthermore, nurses should actively encourage patients to participate in activities that promote patient safety.