Abstract
The use of probiotics to support general health is widespread and quite a bit of research has been conducted on the use of various strains and doses in various contexts. Based on available research, some guidelines have been developed on clinical use of beneficial bacteria under specific disease or treatment circumstances, regarding strains, doses, and expected outcomes.1,2 This article highlights available evidence related to the use of probiotics in abdominal surgery as an addition to standard perioperative procedures, such as prophylactic antibiotic administration, with special attention to the use of probiotics in the perioperative gastrointestinal (GI) oncology setting, a narrow focus that lies beyond the scope of current guidelines.
Introduction
It has become common practice to use probiotics alongside antibiotics, as research suggests they help reduce antibiotic-associated diarrhea and Clostridium difficile-associated diarrhea (CDAD). A meta-analysis of 63 randomized controlled trials (RCTs) totaling 11,811 patients who were treated with antibiotics showed significant association between probiotic administration and a reduction in antibiotic-associated diarrhea (P<0.001).3 Results of a 2013 Cochrane review of probiotics for CDAD suggested that probiotic administration concurrent with antibiotics reduces risk of CDAD by 64% and reduces overall risk of developing side effects.4 However, questions arise as to how broadly this information can be applied clinically and what safety concerns should be considered in various circumstances. Fortunately, research continues to elucidate the limits and scope of appropriate probiotic use, allowing us to hone in on their potential for use alongside antibiotics in specific populations undergoing conventional treatments, such as patients with GI cancer who are undergoing abdominal resection.
At this time, evidence suggests a high likelihood of safety and benefit for perioperative probiotic use in resection of GI malignancies.
GI surgery, along with necessary prophylactic antibiotic administration surrounding the procedure, imposes an insult to the beneficial flora of the intestines and to gut barrier function, which together may contribute to decreased postoperative immune and GI function.5 Failure of gut barrier function and subsequent translocation of endogenous bacteria resulting in systemic inflammatory response is thought to be a major cause of infectious complications following abdominal surgery.5 Additionally, emerging research indicates that endogenous microbes are able to express a virulent phenotype, causing greater tendency for host invasion in response to the surgically induced stress and stress of standard perioperative interventions in the host environment.5 It is thought that probiotics, live beneficial microbes, serve to restore intestinal permeability, maintain intestinal microbial balance, modulate the immune system, and ameliorate the inflammatory response.5 Prebiotics, often administered with probiotics, are fibers that promote growth of beneficial microbes. They reach the colon undigested and are then fermented by beneficial bacteria into important nutrients.5
Probiotic Use in General Abdominal Surgery
The use of perioperative probiotics (or synbiotics when combined with prebiotics) in general abdominal surgery (including but not limited to cancer-related resections) has been shown to improve a variety of outcomes. A 2011 review of prophylactic probiotic use for postoperative infections in 14 RCTs showed as much as a three-fold reduction in postoperative infections for patients who underwent liver transplantation or upper GI tract elective surgery; however, the review did not show reduced postoperative infections in colorectal surgery.6 In this review, the authors suggest differences in outcomes between upper and lower GI procedures may be due to immunomodulating effects of the probiotics, given that the small bowel and liver harbor a significant portion of the immune system. Additionally, because smaller populations of microbes are present in these areas compared to the lower GI tract, it may be easier to manipulate microbial balance. They suggest that a much higher dose and longer duration of probiotics may be needed to reach the lower GI tract to positively impact its microbiota. A 2013 meta-analysis of 13 RCTs totaling 962 patients who underwent elective general abdominal surgery showed reduced incidence of postoperative sepsis in patients taking probiotics and synbiotics vs the control group (P=0.003 and P=0.002, respectively). Some of the studies also showed improved gastric emptying, decreased cramping and distention, decreased diarrhea, and shorter time to first bowel motion.7 In a more recent 2016 systematic review and meta-analysis of 20 RCTs totaling 1,374 patients who underwent elective abdominal surgery, significant reductions in surgical site infections (relative risk [RR]:0.63; 95% confidence interval [CI]: 0.41-0.98), urinary tract infections (RR: 0.29; 95% CI: 0.15-0.57), and combined infections (RR: 0.49; 95% CI: 0.35-0.70) was demonstrated in patients taking probiotics vs control groups.8
The following tables provide an overview of how prebiotics and probiotics have been used in various clinical trials related to surgical resection of abdominal malignancies (Tables 1, 2). The details of how probiotics have been administered in this patient population and the positive outcomes documented may help inform clinical decision-making when considering probiotic use in a patient scheduled to undergo abdominal surgery for a GI malignancy. Total daily amounts are described for easier side-to-side comparison; however, it is worth noting that often the products were administered in divided doses.
| Author, population, study groups | Prebiotic and probiotic strains, doses, route, and schedule of administration | Significant clinical outcomes |
Kanazawa, 20059 N=44 biliary cancer patients undergoing hepatectomy Groups: Enteral feeding plus synbiotics vs enteral feeding only
| Probiotics (6 billion CFUs daily):
Prebiotics:
Administered enterally from postoperative day 1 through day 14 | Incidence of postoperative infections was 19% in the synbiotics group vs 52% in the control (P<0.05) |
Sugawara, 200610 N=81 patients with biliary cancer undergoing hepatectomy Groups: Synbiotics administered preoperatively-only vs synbiotics administered preoperatively and postoperatively
| Preoperative probiotics (50 billion CFUs daily):
Preoperative prebiotics:
Preoperative synbiotics administered orally for 2 weeks prior to hepatectomy Postoperative probiotics (600 million CFUs daily):
Postoperative prebiotics:
Postoperative synbiotics administered enterally from postoperative day 1 through day 14 | Incidence of postoperative infections was 12.1% in preoperatively and postoperatively group vs 30% in preoperative-only group (P<0.05). Duration of antibiotic therapy was shorter in the preoperatively and postoperatively group vs preoperative-only group (media 8 days vs 12 days, P=0.0355) Median length of hospital stay was shorter in preoperatively and postoperatively group vs preoperative-only group (30 days vs 38 days, P=0.0447) |
Rayes, 200711 N=80 patients with either pancreatitis or carcinoma undergoing pancreaticoduodenectomy Groups: Synbiotics vs prebiotics only | Probiotics consisted of 20 billion CFUs blend:
Prebiotics (20 g/d):
Administered daily, orally, or enterally, from postoperative day 1 through day 8 | Incidence of postoperative bacterial infections was 12.5% in the synbiotics group vs 40% in the prebiotics-only group (P=0.005) Duration of antibiotic therapy was shorter in the synbiotics group vs the prebiotics-only group (2+5 days vs 10+14 days, P=0.015) |
Usami, 201112 N=61 patients with hepatic cancer undergoing hepatectomy Groups: Synbiotics vs standard of care | Probiotics (600 million CFUs daily):
Prebiotics:
Administered orally from 14 days preoperatively through postoperative day 11 | Incidence of postoperative infections was zero in the synbiotics group vs 17.2% in the control (P<0.05) |
Sommacal, 201513 N=46 patients with periampullary neoplasms undergoing surgical resection Groups: Synbiotics vs placebo
| Probiotics (8 billion CFUs daily):
Prebiotics:
Administered orally daily for 14 days (4 days preoperatively and 10 days postoperatively) | Incidence of postoperative infections was 26.1% in the synbiotics group vs 69.6% in the control (P=0.00) Duration of antibiotic therapy was shorter in the synbiotics group vs control (9 days vs 15 days, P=0.01) Noninfectious complications were less common in the synbiotics group vs control (6/23 vs 14/23 patients, P=0.03) Mean length of hospital stay was shorter in synbiotics group vs control (12+5 days vs 23+14 days, P=0.00) Zero deaths occurred in the synbiotic group vs 6 deaths in the control group (P=0.02) |
Liu, 201514 N=134 patients with colorectal liver metastasis undergoing resection of primary tumor and liver metastasis Groups: Probiotics vs placebo
| Probiotics (440 billion CFUs daily): 2x1011 CFUs Lactobacillus paracasei 14x1010 CFUs Lactobacillus acidophilus-11 1x1011 Bifidobacterium longum-88 Administered orally for 6 days preoperatively and 10 days postoperatively | Improved outcomes seen in the probiotic group vs the control group, such as septicemia incidence (59% vs 88%, P=0.008), urinary infection (2% vs 13%, P=0.017), diarrhea incidence (24% vs 46%, P=0.012), cumulative duration of antibiotic therapy (6.22 ± 1.96 vs 7.56 ± 2.26, P< 0.001), postoperative hospital stay (11.26 ± 2.52 vs 12.96 ± 3.06, P<0.001) |
Abbreviation: CFU, colony-forming unit
| Author, population, study groups | Prebiotic and probiotic strains, doses, route and schedule of administration | Significant clinical outcomes |
Liu, 201115 N=100 patients with colorectal carcinoma undergoing colectomy Probiotics vs placebo
| Probiotics (440 billion CFUs daily):
Administered orally 6 days preoperatively and 10 days postoperatively | Reduced central lines infection, pneumonia, urinary tract infection, time to defecation, diarrhea incidence, abdominal cramping, abdominal distention, duration of postoperative pyrexia and duration of antibiotic therapy in the probiotic group vs control (P<0.05) |
Zhang, 201216 N=60 patients with colorectal cancer undergoing colorectal resection Probiotics vs placebo
| Probiotics (900 million CFUs daily):
Administered orally for 3 days preoperatively (Days −5 to −3) | Total incidence of postoperative infection was 10% vs 33.3% in groups A and B respectively (P<0.028) with statistically significant differences in bacteremia and septicemia. |
Aisu 201517 N=156 patients with colorectal cancer undergoing resection Probiotics vs standard of care | Probiotics (CFUs not specified):
Administered orally daily starting 3 to 15 days preoperatively, then resumed postoperatively once patient resumed oral water intake | Incidence of superficial surgical site infections was 6.7% in the probiotic arm vs 19.8% in the standard of care arm (P=0.016) |
Kotzampassi, 201518 N=168 Patients with colorectal cancer undergoing resection Probiotics vs placebo | Preoperative probiotics (22 billion CFUs daily):
Postoperative probiotics (11 billion CFUs daily):
Administered daily starting from the day before surgery through postoperative day 14 Capsules were administered orally; in cases with postsurgical nasogastric tubes remaining contents were administered through tube | Significant decrease in major postoperative complications seen in probiotics group vs placebo (28.6% vs 48.8%, P=0.010), including reductions in infectious complications (11.9% vs 28.7%, P=0.009), rate of postoperative pneumonia (2.4% vs 11.3%, P=0.029), surgical site infections (7.1% vs 20.0%, P=0.020), and anastomotic leakage (1.2% vs 8.8%, P=0.031). Time to first major complication was shorter in the placebo group vs probiotic group (P=0.023). Time to first infectious complication was shorter in the placebo group vs probiotic group (P=0.009). Time to first bowel movement was shorter in the probiotic group vs placebo group (P<0.0001). Time to alive hospital discharge was shorter in probiotic group vs placebo group (P<0.0001). |
Yang Y, 201619 N=60 Patients with colorectal cancer undergoing resection Probiotics vs placebo
| Probiotics (180 million CFUs daily):
Administered orally for 5 days preoperatively and 7 days postoperatively | Improved time to first flatus and time to first defecation in probiotic group vs placebo (P=0.03). Decreased incidence of diarrhea in probiotic group vs placebo (P=0.04). |
Abbreviation: CFU, colony-forming unit
Discussion
Though several statistically significant positive outcomes have been documented in various clinical trials on the perioperative use of probiotics in abdominal surgery, in many instances one trial has found no difference in a certain outcome between probiotics and control groups where another trial has found benefit. Conflicting findings could be due to several factors. Many of the trials have included small numbers of participants, which may have made it more difficult for outcomes to reach statistical significance. Additionally, the problem of heterogeneity is most challenging in the evaluation of probiotic studies because the strains, doses, and duration of therapy vary widely, which very likely contributes to conflicting findings. Several studies also use poor descriptors of strains and doses used, further limiting the ability for in-depth analysis of the efficacy of specific strains and doses. Interestingly, the studies of colorectal cancer (Table 2) used probiotics only and did not include use of prebiotics, while all except one of the studies involving upper GI tract cancer resections used synbiotics. Kanazawa et al propose the lack of positive results in some instances may be due to use of probiotics only, without addition of prebiotics.9 Lastly, abdominal surgical procedures vary in degree of invasiveness. In some populations studied, less invasive procedures may have been performed and risk factors for infectious outcomes may have been fewer; in these cases, use of probiotics or synbiotics may not have a significant degree of benefit.
A primary concern regarding the safety of probiotic use in abdominal surgery is risk of systemic infection by the organisms administered. In cancer patients, sepsis secondary to probiotic use is especially a concern due to higher rates of immunosuppression and neutropenia. Saccharomyces boulardii, a variant of Saccharomyces cerevisiae, is generally considered a nonpathogenic yeast. Research supports its use alongside antibiotics to decrease risk of diarrhea and C difficile infection; therefore, it is often used in probiotic formulations.20 Saccharomyces cerevisiae, commonly known as “baker’s” or “brewer’s” yeast, is often used to make bread, beer, and wine. Cases of S cerevisiae fungemia (SF) have been documented in both immunosuppressed21-24 and relatively healthy individuals.23-25 Cases of SF have also been documented after known administration of S boulardii,23 after known ingestion of baker’s yeast,26,27 and in the absence of prior consumption of S boulardii or any significant amount of fermented foods, beverages, or breads.24,28 Estimates indicate the incidence of SF may be 0.1% to 3.6% of all fungemia episodes.23 It is thought that etiology of SF may be either intestinal wall translocation or contamination of vascular access devices by the hands of healthcare workers, and immunocompromised patients may be at higher risk.23,24
Research in Finland analyzed the incidence of Lactobacillus bacteremia (LB) among the general population during an 11-year period when the country experienced a significant increase of Lactobacillus rhamnosus GG consumption.29 Results showed the proportion of Lactobacillus isolates among all blood cultures performed (mean 0.02%) and among all positive blood cultures (mean 0.2%) remained stable over the study period despite a significant increase in L rhamnosus GG consumption. In a 9-year retrospective analysis of LB in patients presenting at a university hospital in France,30 6.6% of the 543,765 blood culture bottles analyzed (121 bottles collected from a total of 38 patients) were positive for bacteria, and of those 0.34% demonstrated a Lactobacillus species. The study found that the most frequent risk factors for LB among LB-positive patients were cancer (40%), immunosuppression (37%), and implanted central venous device (29%). Thirty-one (82%) of LB-positive patients had 1 of those risk factors and 15 (39%) had 2 or more. The authors also reported that 56% had other bacteria present in their cultures in addition to Lactobacillus. The authors hypothesize that LB, with its low virulence, could actually indicate a severe status, poor prognosis, or other comorbidities such as more serious coinfections. They suggest that LB may not necessarily be pathogenic in these cases; rather, its presence should lead to further investigation for the presence of virulent pathogenic microbes or a severe undiagnosed disease process such as an advanced cancer.
In the 2013 meta-analysis of 13 RCTs totaling 962 patients who underwent elective general abdominal surgery, Kinross et al found that none of the reported cases of sepsis were attributed to a probiotic organism.7 In the 2016 systematic review and meta-analysis of 20 RCTs totaling 1,374 patients who underwent elective abdominal surgery, Lytvyn et al found no difference in the incidence of adverse events between probiotic groups and controls, and no reports of serious adverse events attributed to prebiotics or probiotics, including bacteremia or fungemia.8 However, the authors did note that reporting of adverse events was generally poor. More research is certainly needed that can provide clear and diligent documentation of adverse events, routine culturing of sepsis for identification of any potential probiotic species, and more conclusive investigations into the implications of their presence.
A well-designed study in 2008 by the Dutch Acute Pancreatitis Study Group raised concerns of increased risk of acute ischemic bowel disease (resulting in death) when 10 billion CFUs of probiotics were administered prophylactically in divided doses daily alongside standard of care in the treatment of 152 patients with predicted severe acute pancreatitis (SAP) vs use of placebo alongside standard of care in the treatment of 144 patient with predicted SAP.31 The authors proposed probiotics administered enterally may have resulted in increased distention and intraluminal pressure thereby inhibiting proper blood flow through already hypoperfused and ischemic tissues. An updated meta-analysis from 2014, which included the Dutch study, analyzed 6 RCTs totaling 536 patients with predicted SAP, many of whom underwent surgical intervention as well, and found that there was no difference in clinical outcomes, beneficial or adverse, between probiotics and placebo groups.32 These authors also noted significant heterogeneity in strains, doses, and duration of probiotic use among the trials, which likely contributed to varying outcomes.
Clinical trials evaluating use of probiotics in hospitalized critically ill patients seem to have the most reliable data on safety. Patients with SAP are also often included in the evaluations of critically ill populations. A 2013 meta-analysis of 13 RCTs totaling 1,439 hospitalized critically ill patients determined that probiotic administration reduced the incidence of intensive care unit (ICU)-acquired pneumonia (odds ratio [OR]: 0.58; 95% CI: 0.42-0.79) and ICU length of stay (−1.49 days; 95% CI: −2.12 to −0.87 days), with no infections attributed to a probiotic strain and no occurrences of ischemic bowel disease.33 The Dutch study was obviously not included in that meta-analysis but other trials that included SAP cases were. In a more recent 2016 meta-analysis of 30 trials totaling 2,972 hospitalized critically ill patients that did include the Dutch study, the authors found probiotics were associated with significant reduction in infectious complications, including ICU-acquired infections such as ventilator-assisted pneumonia.34 These authors also observed that as newer, quality data emerges there is less heterogeneity of this effect in the trials. They did not describe reports of ischemic bowel disease outside of the Dutch study. Finally, a 2011 review conducted by the American Healthcare Research and Quality agency (AHRQ), which included over 600 clinical trials and case reports on the use of probiotics in both adults and children, healthy or ill, found no association of probiotics with increased risk of adverse events, including infections.35
So, what can we gather from this information to help inform clinical practice in cancer supportive care? At this time, evidence suggests a high likelihood of safety and benefit for perioperative probiotic use in resection of GI malignancies. However, in cases of predicted SAP or significant immunosuppression, perioperative probiotics may not be beneficial and may even be harmful. Until further data on safety emerges in these specific populations, perioperative probiotics are best avoided. In general, it appears from available evidence that it may be well worthwhile to use synbiotics over probiotics alone in the perioperative setting for abdominal malignancies. All but 1 of the studies on synbiotics used a daily dose between 10 and 20 grams. While a wide variety of probiotic strains have been used in studies for GI tract cancer patients undergoing resections, nearly all studies used 4 or fewer strains, often including a combination of Lactobacillus and Bifidobacterium strains. Only 1 study also included S boulardii in their probiotic blend.18 Recall that S boulardii may be beneficial when used alongside antibiotics in decreasing risk of diarrhea and C difficile infection.20 Patients who may be at higher risk of postoperative C difficile infection include those with a history of C difficile infection, those who have a more recent history of extensive antibiotic use, those who have received neoadjuvant chemotherapy or radiation, and especially those who received antibiotics during any portion of their neoadjuvant treatment.36 A synbiotic blend that includes a combination of Lactobacillus strains, Bifidobacterium strains, and S boulardii may be a good choice for perioperative use in this patient population. With such a wide range of doses used in the studies, it may be strategic to use up to 50 billion CFUs orally daily (in divided doses), the typical dose range used; it is unclear at this time if higher doses would have any further added benefit. One might consider dosing on the higher end of that range in lower GI tract resections to ensure sufficient delivery of beneficial bacteria to the colon in order to positively impact its microbiota. Duration of administration often lasted approximately 2 weeks, on average, surrounding the time of the procedure.
An important consideration is the postoperative administration. Since patients are admitted during at least a portion of this time, approval and even an order by the treating surgeon would likely be necessary for oral administration and especially for any enteral administration during hospitalization. One may also consider a lower dose if recommending enteral administration via feeding tube, perhaps a maximum of 20 billion CFUs daily (in divided doses) based on the dose range used in existing studies.
Clinical outcomes of various studies have been the focus of this article, but note that several of these trials also investigated additional parameters such as serum inflammatory and immune markers, fecal presence of both pathological and beneficial microbiota, and various nutritional indicators. What may also be of value for future studies to examine would be psychological differences between groups with perioperative use of synbiotics compared to controls. Limited data exist on the psychological impact of perioperative use of synbiotics or probiotics. In a small trial, 20 patients with laryngeal cancer scheduled to undergo surgery were randomized to either placebo or probiotics and compared to 10 healthy patients who took probiotics.37 The cancer patients received probiotics or placebo daily for 2 weeks, preoperatively (day −14 through day −1). By day 10 of therapy, serum corticotropin-releasing factor and heart rate increased significantly in the placebo group, while both of the probiotic groups remained stable (P<0.05). On day 14, compared to baseline levels, anxiety increased in the placebo group (P<0.05) but decreased in the cancer probiotic group (P<0.05). The healthy group’s anxiety levels were low at baseline and remained unchanged.
With the emergence of more and more research on the relationship of the intestinal microbiome with mental and emotional states outside the surgical context,38 supporting this relationship within the surgical context could be of particular value to patients undergoing major abdominal procedures who are experiencing heightened anxiety or who are at higher risk for anxiety.
Conclusion
In an effort to improve patient outcomes and quality of care in abdominal surgery patients postoperatively, synbiotic use has been shown to be a reasonable, safe, and cost-effective consideration.8 With implementation of the Hospital-Acquired Condition Reduction Program by the Centers for Medicare & Medicaid Services, hospitals may also find financial incentive to take a closer look at perioperative use of synbiotics for reducing infection rates. More research is needed to better describe the degree of efficacy, the impact of various strains, optimal doses, and duration of therapy in various types of cases. In the meantime, we should strive to optimize surgical outcomes using the research currently available and give our patients, especially those at higher risk of postoperative infectious complications, a chance at as smooth a postoperative course as possible.
