Role of aggressive/vigorous hydration with Ringer lactate

Tables 1, 2, and 3 summarize the main details of the trials discussed below.

The first prospective multicenter RCT on the efficacy of aggressive peri-procedural hydration with LR in reducing the incidence of PEP was published by Buxbaum et al. [30]. Although limited by the small number of patients, the study remains a milestone in the field and provides the basis for further studies. In a subsequent prospective, multicenter, double-blind Korean RCT [31], 510 patients at low and high risk of PEP were randomized 1:1 to two groups. The study showed a benefit in the vigorous hydration group. Two similarly designed prospective, randomized, double-blind trials by the same Iranian group, with 150 and 240 patients randomized 1:1, also had similar results [32, 33].

In another Korean double-blind RCT [34] comparing aggressive hydration with LR, aggressive hydration with normal saline, and standard hydration with LR, aggressive hydration with LR reduced the risk of PEP in moderate-to-high risk patients compared to standard hydration with LR. In addition, there were no statistically significant differences in PEP rates between the aggressive normal saline infusion group and the standard LR infusion group, suggesting that aggressive hydration is effective only when performed with LR solution. Furthermore, the need for a bolus immediately after ERCP was demonstrated in a trial where patients randomized to aggressive hydration did not receive any bolus [35].

Finally, two trials demonstrated the non-inferiority of aggressive hydration with LR compared to standard prophylaxis, (i.e., endorectal indomethacin) for the prevention of PEP [36, 37].

Meta-analyses have confirmed the protective role of aggressive hydration in the prophylaxis of post-ERCP acute pancreatitis and hyperamylasemia [38–42], even in low-risk patients [38]. Based on these findings, American guidelines suggest the prophylactic use of aggressive hydration with LR in unselected patients undergoing ERCP [15], while European guidelines recommend its use in patients in whom NSAIDs cannot be administered [13].

Despite such impressive results in well-designed prospective and RCTs, some criticisms can still be made. The first issue concerns the different protocols of aggressive fluid administration used in the above-mentioned trials. The so-called Buxbaum protocol, derived from previous experience in the treatment of acute pancreatitis, was used in the majority of trials [30, 32, 34]. Other trials reported differences in the bolus [31, 35] or lack of hydration during ERCP [33, 36, 37]. These differences are obviously manifested in the different volumes of LR received and the variable gap between aggressive and standard hydration volumes in the different trials. The effects on PEP prevention of aggressive hydration with LR may be affected by differences in the volume of fluid administered. In the study by Chang et al. [35], the difference between the mean total volume infused in the LR group and the standard hydration group was less than 1,200 mL on average (3,600 vs. 2,413 mL). Only two other trials reported the volume of fluid infused. In the study by Choi et al. [31], the volume of fluid administered was 2,744 ± 364 mL and 741 ± 63 mL (p < 0.001) in the vigorous and standard groups, respectively. In the study by Park et al [34], the volume of fluid administered during the 24-hour period was significantly greater in both the aggressive LR and the aggressive normal saline solution groups (4,183 ± 133 mL and 4,180 ± 117 mL, respectively) compared to the standard LR solution group (2,220 ± 69 mL; p < 0.001). According to the meta-analysis by Wang et al. [39], the hydration regimen with the highest efficacy of PEP prophylaxis is reported in the study by Buxbaum et al. [30].

One of the two studies comparing aggressive hydration with indomethacin is also open to criticism because the protocol of administering two doses of indomethacin, 50 mg before and 50 mg after ERCP, contradicts guidelines that recommend a single dose of 100 mg before ERCP [36]. This may explain the high PEP rate in the indomethacin group, which was closer to the placebo group than to the LR group.

In principle, the sample size and the heterogeneity of the method of sample size calculation could be considered as a critical issue. However, when evaluating the individual published trials, the size of the study population per se is not a major issue, nor is the pre-trial power calculation, because, for example, trials may be stopped for ethical reasons, resulting in smaller patient populations.

A comparison with standard hydration is no longer ethically justifiable, as it would be equivalent to comparing an effective prophylactic therapy to placebo, thus exposing some of the enrolled patients to unjustified health risks. There is also limited interest in comparing aggressive hydration with LR to indomethacin alone. Indomethacin has been shown to be effective [11, 12, 43, 44] with a negligible risk of adverse events. Further justification is provided by the trials shown below, which compared LR plus indomethacin to indomethacin alone to investigate a possible benefit of combination therapy.

Despite these critical points, the trials have considerable strengths. First, the primary outcome, i.e., acute pancreatitis, is identical in all trials. The rate and severity of PEP in patients receiving LR prophylaxis are statistically lower than in patients receiving standard hydration. In fact, two trials found a zero rate of severe acute pancreatitis [30, 37]. These results indirectly suggest that LR prophylaxis is beneficial in terms of costs and economic sustainability, by reducing the length of hospital stay and the additional costs associated with managing complications.

Concerning age, the samples and the different groups in each individual study are essentially comparable with regard to the risk of PEP, ranging from a minimum average of 43 years to a maximum of 59 years (Table 1). There were no major differences in sex distribution of the groups among the various trials. With regard to the patient-related risk classification for PEP, the majority of patients were at low and intermediate risk rather than at intermediate and high-risk (1,134 vs. 905). Regarding the presence or absence of intraprocedural technical risks, wherever they were reported, they appeared to be evenly distributed and therefore not able to affect the findings (Table 5). In this connection, even the difficult-to-interpret data from the Park et al. [34] study, which shows very high rates of difficult cannulation, pancreatic duct injection, or recourse to pre-cut, reinforces the idea that aggressive hydration is effective in highly complicated patients due to the combined presence of patient-related and procedure-related risk factors. Furthermore, while operator experience is a documented risk factor for PEP [45], it is noteworthy that the efficacy of aggressive fluid replacement with LR was invariably confirmed irrespective of whether trainees were involved [30, 37] or not [31, 34], and whether the study was single-center or multi-center. Finally, the exclusion of older patients with a higher risk of fluid overload probably resulted in the inclusion of individuals with a low risk of PEP [46], explaining the seemingly paradoxical finding that statistically significant results were obtained with a smaller patient population.

In conclusion, the current evidence presented in our review confirms the role of LR as a substitute for indomethacin in patients with contraindications to the use of NSAIDs, as reiterated in the guidelines.

Role of combined prophylaxis with indomethacin and lactated Ringer

The combined administration of endorectal indomethacin plus aggressive IV hydration with LR is reasonable because the two strategies, which have different mechanisms of action, may have synergistic effects. This combined prophylactic strategy has been investigated in 5 trials, summarized in Tables 4, 5, and 6 [47–51].

In the first single-center small-sized study, Mok et al. [47] compared the incidence of PEP in four different groups of subjects at high risk of PEP. Prophylaxis conducted with indomethacin and LR was found to significantly reduce the incidence of PEP only when compared to the normal saline solution plus placebo (Table 4). Conversely, combining indomethacin with LR was no better than either intervention alone (although a trend towards lower PEP and readmission rates was observed).

In a larger Iranian study [48] and in a subsequent Indian RCT [49], combined prophylaxis with aggressive hydration with LR plus diclofenac showed no additional benefit over single prophylaxis strategies such as NSAIDs or aggressive hydration with LR alone. However, a subsequent prospective Portuguese study found significant superiority of the combined prophylaxis [50].

A recent network meta-analysis comparing different PEP prophylaxis strategies postulated that the combination of indomethacin and LR offered higher protection than single treatments [52]. However, the lack of RCTs directly addressing this research question should be pinpointed.

The recent Dutch FLUYT study, which was conducted in 22 hospitals and included a large number of patients at moderate to high risk of PEP, reached different conclusions [51]. The almost identical incidence of PEP in the two prophylaxis arms led the authors to conclude that PEP prophylaxis with NSAIDs combined with aggressive hydration with LR is not justified. Although this study [51] seemingly provides ultimate evidence against the benefit of combined prophylaxis, some clarifications need to be made. The aggressive hydration protocol used in this study, in which no LR was infused during ERCP, differed slightly from the so-called Buxbaum protocol. Another issue related to intra-procedural risk factors may have affected the findings of the study. Approximately 40% of patients enrolled were exposed to unintentional cannulation of the main pancreatic duct. However, given that PSP was left to the discretion of the operator, it was eventually placed in only 6% of patients in both groups (Table 6). Regarding the type of NSAID used in combination therapy, it should be highlighted that the most favorable results were obtained when the combination therapy included endorectal indomethacin [47, 50] and endorectal diclofenac achieved inferior rates of efficacy. Indeed, a network meta-analysis including 55 RCTs with a total of 20 different interventions in 17,062 patients comparing the effectiveness of different strategies to prevent PEP [53] found that the combination of high-volume intravenous LR plus rectal diclofenac 100 mg and the combination of standard-volume intravenous normal saline plus rectal indomethacin 100 mg were both more effective than rectal indomethacin 100 mg alone [53]. However, these combinations were not significantly more effective than diclofenac 100 mg rectally alone. The meta-analysis concluded that it remains to be proven whether different NSAIDs at different doses have different additive benefits when combined with intravenous fluids [53].

While further studies are eagerly awaited, aggressive infusion of LR in patients at low risk of fluid overload should be considered definitely safe. Indeed, as in the trials using prophylactic LR alone, all 5 trials shown in Table 5 showed a negligible rate of serious adverse events.

To summarize, trials have shown that LR is effective in preventing PEP. Further adequately powered studies will clarify whether combination therapy can be a valid option for PEP prophylaxis.