Early Rescue from Acute Severe Clostridium Difficile : A Novel Treatment Strategy

Abstract

Background:

Severe Clostridium difficile infections (CDI) can lead to significant impediments to effective treatment. We developed a novel treatment protocol utilizing bedside gastrointestinal lavage (GIL) for the management of patients with severe, complicated CDI. We describe the development and early outcomes of non-operative bedside GIL in hospitalized patients with severe, complicated CDI following the Idea, Development, Exploration, Assessment, Long Term Study (IDEAL) framework at the Idea stage. We compared our results with those of a cohort of patients managed with colectomy.

Methods:

We conducted a retrospective cohort study of hospitalized patients with severe, complicated CDI who failed conventional medical therapy and were referred for surgical consultation at two academic tertiary-care hospitals between January 2009 and January 2015. After surgical assessment, the attending surgeon decided to proceed either with bedside GIL or directly to colectomy. Bedside GIL involved nasojejunal tube insertion followed by flushing with 8 L of polyethylene glycol 3350/electrolyte solution over 48 h. Both patient groups received standard medical treatment with vancomycin 500 mg q 6 h enterally and metronidazole 500 mg intravenously three times daily for 14 d. The main outcomes of interest were the incidence of colectomy, complications, and mortality rate.

Results:

Nineteen and seventeen patients underwent GIL and direct colectomy, respectively. There were no significant differences between the groups in terms of demographics, American Society of Anesthesiologists class, disease severity, need for intensive care unit admission, mechanical ventilation, vasopressor use, serum lactate concentration, or proportion presenting with hypotension, acute kidney injury, or a white blood cell count >16,000/mcL or <4,000/mcL (p > 0.1). The in-hospital mortality rate was 26% (5/19) and 41% (7/17) for the GIL and colectomy groups, respectively (p = 0.35). Only one patient in the GIL group failed the protocol, requiring colectomy. There were no significant differences in complications in the two groups.

Conclusions:

Bedside GIL appeared to be safe for the treatment of patients with severe, complicated CDI who had failed conventional medical therapy. It did not appear to increase the risk of morbidity or death compared with the traditional strategy of proceeding directly to colectomy.

The incidence, burden of illness, complications, and mortality rate associated with Clostridium difficile infections (CDI) has increased over the past decade [1 –6]. In 2011, almost 500,000 CDI and 29,300 associated deaths were identified in the United States [7]. The spectrum of CDI ranges from uncomplicated diarrhea to severe colitis and multi-organ dysfunction leading to death [8]. Treatment recommendations are based on disease severity, history of CDI, and risk of recurrence [9 –11]. Although the majority of CDI are treated successfully with oral antibiotics, as many as 37% of patients will manifest severe or complicated disease with significant morbidity and a high mortality rate [12]. Treatment guidelines recommend that patients with severe, complicated disease (indicated by megacolon or hemodynamic instability) undergo colectomy [9 –11]. Prompt surgical intervention in such cases has been associated with better survival rates [13], yet the mortality rate after surgery remains devastating highly at 31% to 80% [14 –16].

More recently, intra-operative colonic lavage via loop-ileostomy has shown promising results [17]. However, the safety and efficacy of this approach, especially in frail patients, has not been established clearly, thus the impetus for devising a less-invasive lavage protocol to facilitate earlier aggressive treatment. We developed a non-operative protocol of bedside gastrointestinal lavage (GIL) via nasojejunal (NJ) tube. The primary objective of this study was to describe the development of this therapeutic innovation of using non-operative GIL in hospitalized patients with severe, complicated CDI. Our secondary objective was to assess the safety of this method compared with the traditional approach of immediate colectomy.

Patients and Methods

We conducted a retrospective cohort study of hospitalized patients with severe, complicated CDI who were refractory to medical management, defined as persistent or worsening disease despite maximum medical therapy, who were referred for surgical consultation. Worsening disease was marked by any of the following: (1) rising white blood cell (WBC) count; (2) increasing abdominal pain or distension; (3) radiographic evidence of increasing small bowel or colonic distension or bowel-wall thickening; or (4) evidence of end-organ dysfunction (hemodynamic instability, need for mechanical ventilation, serum creatinine elevation to 1.5 × the baseline value, oliguria/anuria, elevated serum lactate concentration, or metabolic acidosis). Clostridium difficile infection was classified as severe or severe–complicated according to standard definitions [9,10]. After assessment, surgeons decided whether to perform bedside GIL or surgery. They proceeded to colectomy if they deemed a patient had failed to improve with GIL. Patients who underwent GIL were compared with patients who went directly to colectomy over the last five years. Subtotal colectomy was performed in the standard open fashion with creation of an end-ileostomy. Both patient groups continued to receive standard medical treatment with vancomycin 500 mg q 6 h via NJ and metronidazole 500 mg intravenously three times daily for 14 d. We excluded any patient who had evidence of perforation at the time of surgical consultation, as this is an absolute indication for colectomy.

This retrospective analysis was approved by our Institutional Health Sciences Research Ethics Board (HSREB No. 104944). Data were analyzed using the Fisher exact test to compare categorical variables, whereas the Student t-test or Mann-Whitney U test were used to compare continuous variables. A two-sided alpha of 0.05 was used for all tests of significance. Analyses were performed using SPSS/PASW v. 22 (IBM Corporation, Armonk, NY).

Gastrointestinal lavage protocol and development

Bedside GIL has been used at our tertiary-care center since 2012. The protocol consists of: (1) Insertion and radiographic confirmation of proper placement of a small NJ feeding tube; (2) insertion of a fecal management system (Flexiseal^™; ConvaTec, Skillman, NJ); and (3) gastrointestinal lavage with 8 L of polyethylene glycol 3350/electrolyte solution (PegLyte Powder; Pharmascience Inc., Québec) over 48 h in 4-L aliquots. We instituted the fecal management system to track output in response to the high-volume lavage in patients with potential ileus in order to reduce the risk of regurgitation and subsequent aspiration. During the development of this protocol, concern was raised about the potential for aspiration secondary to the gastric position of the NJ tube. Thus, the protocol was refined to require confirmation of post-pyloric positioning of the feeding tube.

Severity scoring systems

Given the lack of consensus on the utility of CDI severity scoring systems, we elected to use two systems to assess and compare disease severity between the GIL and colectomy groups. The first system, proposed by Neal et al. in the paper describing their results using intra-operative colonic lavage via loop-ileostomy [17], defines CDI severity at diagnosis and was not derived for prediction of outcomes. There are 12 indices, and each index carries a point score [17]. One to three points is defined as mild to moderate disease, four to six points as severe disease, and seven or more points as severe, complicated disease. Three of the indices (vasopressor requirement, mechanical ventilation, and altered mental status) are assigned a score of five points, which places patients immediately in the severe disease category [17]. We also used the system developed by van der Wilden et al., which has four variables (age, WBC count, cardiorespiratory failure, and abdominal tenderness) [18]. This system was developed and validated to predict development of fulminant C. difficile colitis. A score of six or greater is classified as high risk of fulminant disease.

Results

We identified 19 patients who underwent GIL and 17 patients who proceeded directly to colectomy. There was no significant difference between the groups in age, sex, American Society of Anesthesiologists (ASA) class, immunosuppression, need for intensive care unit (ICU) admission (for mechanical ventilation or vasopressor use), or proportion with hypotension or tachycardia (Table 1). There were no significant differences in the mean serum lactate or creatinine concentrations or WBC count at the time of surgical consultation. Pre-admission independent functional status was used as a measure of frailty, with no significant difference identified between the groups (p = 0.52). There were more patients with recurrent CDI and fewer patients with peritonitis in the GIL group than in the colectomy group. The GIL group had lower disease severity scores according to the systems proposed by Neal et al. and Van der Wilden et al.; however, this difference was not statistically significant on parametric or non-parametric testing. The GIL group scores were still in the severe and high-risk ranges [17,18]. One patient in the GIL group failed treatment and eventually required colectomy (Table 2). This failure was recognized early in the course of utilizing GIL (in the initial 24 h of the lavage protocol), and the patient had immediate colectomy.

Table 1.

Patient Baseline Characteristics Stratified by Intervention

Characteristic	Gastrointestinal lavage (n = 19)	Colectomy (n = 17)	p
Mean age, y (SD)	68.0 (14.9)	65.4 (16.9)	0.63
Female (%)	10 (52.6)	7 (41.2)	0.52
ASA Class (%)			0.54
2	2 (10.5)	3 (17.6)
3	11 (57.9)	11 (64.7)
4	6 (31.6)	3 (17.6)
Independent functional status (%)	10 (52.6)	11 (64.7)	0.52
Mean leukocyte count at diagnosis × 10⁹/L (SD)	21.0 (18.0)	21.3 (15.6)	0.96
Mean leukocyte count at surgical consultation × 10⁹/L (SD)	21.1 (15.3)	27.0 (19.0)	0.31
Leukocyte count >16.0 or <4.0 × 10⁹/L (%)	8 (42.1)	11 (64.7)	0.18
Mean serum lactate concentration at surgical consultation mmol/L (SD)	2.0 (1.3)	3.0 (2.6)	0.15
ICU admission (%)	8 (42.1)	7 (41.2)	0.95
Vasopressor required	7 (36.8)	6 (35.3)	0.92
Mechanical ventilation	2 (10.5)	3 (17.6)	0.54
Immunosuppression (%)	10 (52.6)	7 (41.2)	0.49
Recurrent disease (%)	11 (57.9)	4 (17.6)	0.04
Peritoneal signs (%)	1 (5.3)	6 (35.3)	0.02
Abdominal distention (%)	10 (52.6)	14 (82.3)	0.06
Mental status changes (%)	3 (23.1)	7 (41.2)	0.09
Acute renal dysfunction (%)	7 (36.8)	11 (64.7)	0.10
Systolic blood pressure <90 mm Hg (%)	9 (47.4)	7 (41.2)	0.71
Heart rate >100 bpm (%)	10 (52.6)	12 (70.6)	0.27
Disease severity score
Mean Neal (SD)	9.5 (6.3)	12.6 (5.3)	0.56
Median Neal (IQR)	7.0 (2.0, 8.5)	13.0 (8.0, 15.5)	0.11
Mean van der Wilden (SD)	6.3 (4.0)	9.0 (4.6)	0.07
Median van der Wilden (IQR)	7.0 (4.5, 12.5)	8.0 (7.0, 13.5)	0.11

ASA = American Society of Anesthesiologists; IQR = interquartile range; SD = standard deviation.

Table 2.

Patient Outcomes Stratified by Bedside Gastrointestinal Lavage or Colectomy for Treatment of Acute Severe Clostridium difficile Infections

Outcome (%)	Gastrointestinal lavage (n = 19)	Colectomy (n = 17)	P
In-hospital death	5 (26.3)	7 (41.2)	0.35
Colectomy	1 (5.3)	N/A	N/A
Venous thromboembolism	2 (10.5)	1 (5.9)	0.62
Aspiration pneumonia	1 (5.3)	2 (11.8)	0.48
Pneumonia	3 (15.9)	9 (52.9)	0.06

The in-hospital mortality rate was 26% for the GIL group and 41% for the colectomy groups, but this difference was not statistically significant (p = 0.35). There were no significant differences in complications, including venous thromboembolism, aspiration, and pneumonia (Table 2).

Discussion

Total abdominal colectomy is currently recommended as the surgical intervention of choice in patients with severe or complicated CDI [9,15,19]. Early surgical intervention for severe or complicated CDI, defined as surgery prior to the development of shock or requirement for vasopressors, is recommended because of the potential for significant mortality reduction (risk reduction [RR] 0.50; 95% confidence interval [CI] 0.35–0.72) [20]. However, the high morbidity and mortality rates associated with surgery and uncertainty about which patients will suffer more fulminant stages of CDI likely contributes to a delay in definitive operative management until the patient is severely deranged physiologically. Although intra-operative colonic lavage via loop-ileostomy has shown promising results, surgeons at our center continue to have the same degree of hesitation to proceed to surgery and perform intra-operative lavage, especially in frail patients [17]. Thus, the time to surgery and the disease severity were essentially the same for patients managed with colectomy or intra-operative colonic lavage in the few who received this treatment (unpublished data). This observation was the impetus for developing a less-invasive gastrointestinal lavage protocol to facilitate earlier aggressive treatment.

This article describes the development and early exploration of this therapeutic innovation. According to the Idea, Development, Exploration, Assessment, Long-term follow-up (IDEAL) framework for the development of surgical innovation, our report falls into stages 2a/b, the Development and early Exploration Stages [21]. Although this is the first report of non-operative GIL for this indication and thus would mean we also are reporting on stage 1 (i.e., Idea), it is an extrapolation of the operative innovation described by Neal et al. [17]. Our findings suggest that GIL via NJ tube is safe for the management of severe CDI. Although there was a concern about the potential for more cases of pneumonia because of the greater aspiration risk with administration of large-volume NJ lavage in patients with CDI, we found no signal of more cases of pneumonia or respiratory failure in the GIL patients compared with those undergoing colectomy.

We hypothesize that the efficacy of bedside GIL in severe, complicated CDI was, in some part, attributable to the non-invasive nature of the protocol and thus, resulted in application of this treatment earlier in the course of the disease with the goal of avoidance of the morbidity and death associated with surgery. We postulate that earlier aggressive intervention is an important factor in improving outcomes. In our study, patients who underwent GIL were less likely to have peritoneal signs and abdominal distention, suggesting that these patients were earlier in the disease course than the colectomy patients. Because the patients receiving GIL appeared to have better outcomes than patients who underwent colectomy, we propose that aggressive management should be instituted early before peritoneal signs and distention occur. Although still in the severe and high-risk categories according to the Neal et al. and van der Wilden et al. scoring systems, the GIL group tended to have lower scores than the colectomy group. Thus, application of aggressive gastrointestinal lavage (via NJ or ileostomy) might be a therapeutic strategy that could be utilized effectively in patients earlier in the course of their disease, as signaled by severity scoring systems. Alternatively, our outcomes might have been better because our patients were not as sick as the patients who underwent colectomy directly. This is certainly a possibility, given the retrospective nature of the data. However, the patients who received our GIL protocol certainly had severe to severe, complicated disease. In general, patients having GIL were perceived to be frailer than the patients who went directly to colectomy. Although there were no differences between the groups in our measure of frailty, the GIL protocol was initially used in our center only in patients who were considered by the surgical team or the patient/family to be too frail to survive surgery. Surgeons were hesitant to delay surgery in patients who were “fit enough” for surgery. In many ways, the inherent selection bias present in retrospective studies is prejudicing the results against the GIL group. This is because most patients received the GIL non-operative therapy precisely because they were deemed to be too frail to tolerate surgery; thus, right from the outset, they were selected into a group with a higher perceived propensity for poor outcomes, including death. We believe that earlier application of aggressive therapy contributed to the better outcomes. However, prospective studies are needed to compare the outcomes of patients randomized between GIL and conventional therapy.

The limitations of this study are those typically found at this stage of the IDEAL framework. Limitations include the retrospective study design and a small sample that may have reduced the power to detect actual differences in baseline characteristics, risk factors, and outcomes. An important limitation is the potential for selection bias, and the two groups may in fact have had different CDI severity. Although we are comparing data on GIL and colectomy, these are likely different patient populations with significant selection bias. We compared them in order to identify key differences that may help clinicians identify patients who may benefit from GIL and for whom it may be safe to try this measure as the first step. Our comparison of the GIL and colectomy patients showed that peritonitis and abdominal distention were significantly more common in the colectomy group. This is certainly a sign that these two groups are different. This may also be an indication that the best time to select patients for GIL may be before peritonitis and abdominal distention occur. Thus, the selection bias may itself be instructive in trying to determine which degree of severity should serve as the target for a less-invasive intervention such as GIL.

In conclusion, GIL was not associated with an increase in morbidity or death compared with the usual care of severe, complicated CDI. There was a signal toward a reduced mortality rate; however, drawing conclusions from these data is tempered by the limitations of the retrospective, non-randomized nature of the study. Further prospective studies are required to assess the efficacy of this treatment. A one-year, single-center pilot randomized controlled trial comparing GIL with medical therapy for patients with severe, complicated CDI without immediate surgical indication is in progress at our institution (ClinicalTrials.gov: NCT02466698) [22]. We expect that a multicenter trial will follow this pilot study.

Footnotes

Acknowledgment and Contributions

T. Mele and B. Kidane had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the analysis. The authors declare that there are no conflicts of interest or any source of funding.

The authors who made substantial contributions to conception and design were BK, DG and, TM. Those who participated in acquisition of data were BK, KL, GM, CE, MO, RH, CV, NG, KL, and TM. Those who analyzed and interpreted the data were BK and TM. All authors participated in drafting or revising the article critically for important intellectual content and gave final approval of the version to be published.

Preliminary work presented at the Annual Meeting of the Surgical Infection Society, Westlake Village, California, April 15–18, 2015, USA. Updated work presented at the Meeting of the Canadian Association of General Surgeon, Québec City, Québec, Canada, September 17–20, 2015.

Author Disclosure Statement

No competing financial interests exist.

References

Kuy

, Jenkins

, Romero

, et al. Increasing incidence of and increased mortality associated with Clostridium difficile-associated megacolon. JAMA Surg, 2016; 151:85–86.

Gravel

, Miller

, Simor

, et al. Health care-associated Clostridium difficile infection in adults admitted to acute care hospitals in Canada: A Canadian Nosocomial Infection Surveillance Program Study. Clin Infect Dis, 2009; 48:568–576.

McDonald

, Owings

, Jernigan

. Clostridium difficile infection in patients discharged from US short-stay hospitals, 1996–2003. Emerg Infect Dis, 2006; 12:409–415.

Redelings

, Sorvillo

, Mascola

. Increase in Clostridium difficile-related mortality rates, United States, 1999–2004. Emerg Infect Dis, 2007; 13:1417–1419.

Steiner

, Barrett

, Weiss

. HCUP projections: Clostridium difficile hospitalizations 2001 to 2013. Rockville, MD: Agency for Healthcare Research and Quality, 2014. Contract No. 2014-01.

Chitnis

, Holzbauer

, Belflower

, et al. Epidemiology of community-associated Clostridium difficile infection, 2009 through 2011. JAMA Intern Med, 2013; 173:1359–1367.

Lessa

, Winston

, McDonald

, et al. Burden of Clostridium difficile infection in the United States. N Engl J Med, 2015; 372:2369–2370.

Rupnik

, Wilcox

, Gerding

. Clostridium difficile infection: New developments in epidemiology and pathogenesis. Nat Rev Microbiol, 2009; 7:526–536.

Cohen

, Gerding

, Johnson

, et al. Clinical practice guidelines for Clostridium difficile infection in adults: 2010 update by the Society for Healthcare Epidemiology of America (SHEA) and the Infectious Diseases Society of America (IDSA). Infect Control Hosp Epidemiol, 2010; 31:431–455.

10.

Debast

, Bauer

, Kuijper

, et al. European Society of Clinical Microbiology and Infectious Diseases: Update of the treatment guidance document for Clostridium difficile infection. Clin Microbiol Infect, 2014; 20(Suppl 2):1–26.

11.

Surawicz

, Brandt

, Binion

, et al. Guidelines for diagnosis, treatment, and prevention of Clostridium difficile infections. Am J Gastroenterol, 2013; 108:478–498.

12.

Levy

, Szabo

, Lozano-Ortega

, et al. Incidence and costs of Clostridium difficile infections in Canada. Open Forum Infect Dis, 2015; 2:ofv076.

13.

Stewart

, Hollenbeak

, Wilson

. Is colectomy for fulminant Clostridium difficile colitis life saving? A systematic review. Colorectal Dis, 2013; 15:798–804.

14.

Halabi

, Nguyen

, Carmichael

, et al. Clostridium difficile colitis in the United States: A decade of trends, outcomes, risk factors for colectomy, and mortality after colectomy. J Am Coll Surg, 2013; 217:802–812.

15.

Abou Chakra

, Pepin

, Sirard

, et al. Risk factors for recurrence, complications and mortality in Clostridium difficile infection: A systematic review. PloS One, 2014; 9:e98400.

16.

Seder

, Villalba

Jr , Robbins

, et al. Early colectomy may be associated with improved survival in fulminant Clostridium difficile colitis: An 8-year experience. Am J Surg, 2009; 197:302–307.

17.

Neal

, Alverdy

, Hall

, et al. Diverting loop ileostomy and colonic lavage: An alternative to total abdominal colectomy for the treatment of severe, complicated Clostridium difficile-associated disease. Ann Surg, 2011; 254:423–427.

18.

van der Wilden

, Chang

, Cropano

, et al. Fulminant Clostridium difficile colitis: Prospective development of a risk scoring system. J Trauma Acute Care Surg, 2014; 76:424–30.

19.

Markelov

, Livert

, Kohli

. Predictors of fatal outcome after colectomy for fulminant Clostridium difficile colitis: A 10-year experience. Am Surg, 2011; 77:977–980.

20.

Ferrada

, Velopulos

, Sultan

, et al. Timing and type of surgical treatment of Clostridium difficile-associated disease: A practice management guideline from the Eastern Association for the Surgery of Trauma. J Trauma Acute Care Surg, 2014; 76:1484–1493.

21.

Pennell

, Hirst

, Campbell

, et al. Practical guide to the Idea, Development and Exploration stages of the IDEAL framework and recommendations. Br J Surg, 2016; 103:607–615.

22.

McCreery

, Jones

, Kidane

, et al. Polyethylene glycol intestinal lavage in addition to usual antibiotic treatment for severe Clostridium difficile colitis: A randomised controlled pilot study. BMJ Open, 2017; 7:e016803.