Tech Coloproctol (2008) 12:21–25 DOI 10.1007/s10151-008-0393-y
O R I G I N A L A RT I C L E
Emergency Hartmann’s procedure: morbidity, mortality and reversal rates among Asians Q.M. Leong • D.C. Koh • C.K. Ho
Received: 20 June 2007 / Accepted: 20 November 2007
Abstract Background Hartmann’s procedure and its reversal are associated with high cumulative morbidity rates. We assessed the outcomes of emergency Hartmann’s procedures and permanent stoma rates among Asians. Methods A retrospective review of all emergency Hartmann’s procedure performed from 1996 to 2001 was performed. Results Emergency Hartmann’s procedure was performed in 98 patients with a median age of 68 years. These included 58 patients with cancer and 18 with diverticulitis; 77 patients had pre-existing comorbidities. The mortality and morbidity rates were 19% and 65%, respectively. Mortality was impacted by pre-existing respiratory or cardiac disease, age and ASA grade. Likewise, morbidity was significantly increased in the presence of comorbidities, including hypertension and cardiac disease,
and age. On multivariate analysis, only age (p=0.003, OR=1.171) and respiratory disease (p=0.029, OR=11.05) affected mortality rates whilst hypertension (p=0.011, OR=5.85) and cardiac disease (p=0.044, OR=5.46) affected morbidity rates. Re-anastomosis was performed in 70 patients, and in 28 of these patients (40%) bowel continuity was re-established after a median of 9 months with a 7% morbidity rate. Reversibility was only related to patient age (p=0.011). Conclusions Hartmann’s procedure is valid among Asians, and its mortality and morbidity rates are related to patients’ pre-existing health conditions. The predominant cause is colorectal cancer and permanent stoma rates are related to patient age. Key words Emergency · Hartmann’s procedure · Asia
Introduction
Q.M. Leong (쾷) · C.K. Ho Department of General Surgery Tan Tock Seng Hospital 11 Jalan Tan Tock Seng, Singapore 308433 e-mail:
[email protected] D.C. Koh Colorectal Surgery Division Department of Surgery National University Hospital, Singapore
In 1921, Henri Hartmann described an operation for the treatment of cancers of the distal sigmoid colon and rectum in which the tumour was resected, the rectal stump closed and an end colostomy fashioned [1]. It is not known whether Hartmann himself ever intended to restore the continuity of the colon, although this is now regarded as the second stage of the procedure. Boyden was the first to describe the use of this procedure in the surgical management of acute diverticulitis [2]. Cancer and diverticulitis remain the main indications for Hartmann’s procedure today. Other indications such as sigmoid volvulus, colorectal anastomotic leak, ischemic colitis and iatrogenic colorectal perforations have also been described [3–6]. Hartmann’s procedure became popular in the 1970s, when it represented an evolution in the treatment of left-sided colonic pathology,
Tech Coloproctol (2008) 12:21–25
22
especially in the emergency setting. It compared favourably to the historical 3-stage procedure (initial colostomy, followed by resection at another setting and finally closure of the stoma). The main advantage of the Hartmann’s procedure is the immediate resection of the diseased bowel segment whilst avoiding the potential complications of a colorectal anastomosis, especially in an emergency setting where the bowel is unprepared. The main disadvantages of Hartmann’s procedure are a low reversal rate of approximately 60% and the associated morbidity of undergoing the reversal operation, quoted to be as high as 34.7% in one study [7]. Major advances made in anaesthesia, antibiotics, intensive care management and surgery in the last few decades, together with the technique of intra-operative on-table lavage, have facilitated safe primary anastomosis to be performed in a one-stage procedure in selected patients, with reported low morbidity and mortality rates [8–11]. The presence of septic shock, faecal peritonitis, immuno-compromised status and American Society of Anesthesiologists grade 4 (ASA 4) are well accepted contraindications to the one-stage procedure. Under these suboptimal conditions, Hartmann’s procedure is still considered a safe and valid option [14]. The primary aim of this study was to determine the mortality and morbidity rates of emergency Hartmann’s procedure amongst Asians. Our secondary aim was to determine if there were factors which affected the rates of stoma reversal.
Materials and methods Between February 1996 and December 2001, a total of 376 colorectal resections were performed in Tan Tock Seng Hospital, a 1500-bed institution. Of these, 98 patients (26.1%) had undergone an emergency Hartmann’s procedure for left-sided colorectal pathology. Data regarding demographics, time to surgery, surgeon’s expertise, American Society of Anesthesiologists (ASA) grade, curative intent, duration of operation, length of stay and postoperative outcomes were extracted from the hospital’s clinical database. Postoperative mortality was defined as death occurring within 30 days of the operation, or during the same hospital admission as a consequence of the operation. Statistical analysis was performed with the chisquared test or Fisher’s exact test for nominal variables. The 2 sample t test was used to compare continuous data from independent samples. Multiple logistic regression was used to analyse the factors associated with mortality and morbidity. All statistical tests were performed using
the Statistical Package for Social Sciences for Windows (SPSS version 12.0); p values less than 0.05 were considered significant.
Results The 98 patients in the analysis included 56 men and 42 women of median age 68 years (range, 30–92). Racial distribution was as follows: 83 Chinese (85%), 11 Malays (11%), 3 Indians (3%) and 1 Eurasian (1%); this distribution was similar to that of Singapore’s multiracial population, comprising 76.8% Chinese, 13.9% Malays, 7.9% Indians and 1.4% other races [15]. The most frequent indications for emergency Hartmann’s procedure were colorectal cancer in 58 patients (59%) and complicated diverticulitis in 18 patients (Table 1). The cause was not clearly documented in 2 patients. Of the 58 patients with cancer, 41 presented with intestinal obstruction while 17 had acute perforation. All patients with diverticular disease presented with perforation: 3 patients had a Hinchey score of 1, 6 patients had a score of 2, another 6 patients had a score of 3, and 3 had a score of 4. Comorbidities were present in 77 patients (79%). These included hypertension (27%), cardiac disease (21%), diabetes mellitus (20%), respiratory disease (10%), previous cerebrovascular accidents (9%), various thyroid disorders (12%) and other medical problems (28%) (Table 2). Many patients had multiple co-morbidities. Twenty one patients (21%) had a pre-operative ASA score of 1, 37 (38%) an ASA score of 2, 30 (31%) an ASA score of 3, and 10 (10%) an ASA score of 4. There were 19 surgical mortalities (19%) in this series. The causes of death are shown in Table 3. Table 1 Indications for emergency Hartmann’s procedure in 98 patients with left-side colorectal pathology Diagnosis Cancer Obstruction Perforation Complicated diverticulitis Anastomotic leak Volvulus Adhesions Ruptured abdominal aortic anerysm Ischaemic colitis Iatrogenic perforation Road Traffic accident Strangulated hernia Ulcer Carcinomatosis peritonei Ovarian mass Unknown
Patients, n (%) 58 (59) 41 (42) 17 (17) 18 (18) 6 1(6) 3 1(3) 2 1(2) 2 1(2) 1 1(1) 1 1(1) 1 1(1) 1 1(1) 1 1(1) 1 1(1) 1 1(1) 2 1(1)
Tech Coloproctol (2008) 12:21–25
23
Statistical analysis showed that the mortality rate was significantly related to 4 factors: pre-existing respiratory disease (p=0.022) and cardiac disease (p=0.002), age (p=0.001) and ASA score (p=0.016). After multiple logistic regression, only the presence of pre-existing respiratory disease (p=0.029; OR=11.05; 95% CI, 1.276 to 95.721) and age (p=0.003; OR=1.171; 95% CI, 1.054 to 1.301) correlated with mortality. The median operating time for the procedure was 165 minutes (range, 60–300 minutes). Sixty two patients (63%) had their procedures performed by consultant surgeons whilst registrars performed the remaining 36. The surgeon’s level of expertise (consultant versus registrar) did not have any significant impact on operative time. Sixtysix (65%) patients in the study developed a postoperative morbidity after the Hartmann’s procedure. Among these patients, 23 (23%) developed pneumonia and 20 (20%) developed wound infections. Some patients had multiple morbidities. The rest of morbidities Table 2 Comorbidities in 98 patients who had emergency Hartmann’s procedure Comorbidity
Patients, n (%)
Hypertension Cardiac disease Diabetes mellitus Respiratory disease Cerebrovascular accidents Thyroid disease Other
27 21 20 10 9 12 28
Table 3 Causes of death after emergency Hartmann’s procedure Cause Pneumonia Septicaemia Heart failure Stroke Small bowel infarct Renal failure Other Total
Patients, n (%) 9 1(47) 3 1(16) 2 1(11) 1 11(5) 1 11(5) 1 11(5) 2 1(11) 19 (100)
Table 4 Morbidity following Hartmann’s procedure Morbidity Pneumonia Wound infection Renal failure Ileus Acute myocardial infarction Atrial fibrillation Cerebrovascular events Burst abdomen Intra-abdominal abscess Other Total
Patients, n (%) 23 1(23) 20 1(20) 9 11(9) 4 11(4) 4 11(4) 3 1v(3) 3 11(3) 2 11(2) 2 11(2) 15 1(18) 85 (100)
are shown in Table 4. The presence of a morbidity correlated significantly with the presence of pre-existing comorbidities (p=0.001), hypertension (p=0.011), cardiac disease (p=0.006) and age (p=0.003). After multiple logistic regression, only the presence of pre-existing cardiac disease (p=0.044; OR=5.462; 95% CI, 1.044 to 28.572) and hypertension (p=0.011; OR=5.848; 95% CI, 1.512 to 22.619) correlated with morbidity. Sixteen of the 98 procedures were palliative in nature. The remaining 82 were intended to be staged procedures, but 12 patients died during the same admission, before re-anastomosis could be planned. Of the 70 who survived, 28 had bowel continuity re-established, giving an overall reversal rate of 40%. On analysis of this subset of patients, 13 (46%) had colorectal cancer, 8 (29%) had complicated diverticular disease, and 7 (25%) had other benign pathology, giving a disease-specific reversal rate of 30% for those with colorectal cancer and 57% for those with complicated diverticular disease. The median time to the reversal procedure was 9 (range, 6–18) months. We found that an increased age was significantly related to a lower likelihood of stoma reversal in our patients (p=0.011). However, ASA grade, hypertension, diabetes mellitus and cardiac or respiratory co-morbidities were all not significant predictors of patients having their stomas reversed. Those with a benign pathology showed a trend towards a higher likelihood of having their stomas reversed (p=0.09). When we looked at the 42 patients who did not undergo the reversal procedures, we found that the reasons for this included high risk due to co-morbidities in 17 patients (40%), cancer-positive resection margins in 7 (17%), patient refusal in 5 (12%) and recurrent disease in 1 (2%). Six patients (14%) were lost to follow-up and 3 (7%) died during follow-up. The reason for non-reversal was not known in 2 patients (5%). Attempted reversal in 1 patient was unsuccessful due to dense adhesions encountered during the pelvic dissection. There were no mortalities associated with the reversal procedure in our series. Two patients (7%) developed postoperative complications. These included an incisional hernia which was treated conservatively, and an anastomotic leak that required a re-laparotomy and permanent colostomy. Both patients were well as of the last follow-up, with no further surgical intervention planned.
Discussion The median age of our patients was 68 years and this is comparable to those reported in other series [3, 5, 6]. The predominant indications for performing emergency Hartmann’s procedure in our patient collective were col-
Tech Coloproctol (2008) 12:21–25
24
orectal carcinoma (59%) and complicated diverticular disease (19%). The higher proportion of cancer as a cause differs from that found in the west, and reflects the spectrum of colorectal diseases found in our population in which left-sided diverticular disease is less common and an increasing incidence of colorectal cancer is being reported [12, 13, 16]. Where malignant disease is present, the Hartmann’s procedure has been considered a useful option in the management of obstruction or perforation at the tumour site, as reflected in our series. The presence of faecal peritonitis or gross purulent peritonitis is often deemed as a contraindication towards performing a colonic anastomosis for fear of an increased likelihood of anastomotic leaks. Hence, 47% of our patients, who presented with complicated diverticulitis, had either faecal or purulent peritonitis, making the Hartmann’s procedure a safe choice. The general pre-operative condition of the patient is also an important consideration in the decision-making process. The presence of systemic factors, such as diabetes, renal failure, cardiovascular disease, persistent hypotension, compromised immunity and malnutrition can jeopardize anastomotic healing [14]. In our series, 79% of our patients had at least one of these co-morbidities, and 41% had an ASA score of 3 or 4. Hence, the Hartmann’s procedure may be the safest option. The median length of stay was 13 days (range, 1–69 days), similar to that reported in most other series [3–6]. This is despite the fact that all cases were performed as an emergency and that the majority of our patients were in a poor state of health. For the first-stage procedure, our mortality of 19% is similar to that of other reported series, although our 65% morbidity is high [4, 5]. The most common morbidities in our series were pneumonia (23%) and wound infection (20%), similar to another series [4]. A likely explanation for our rates of pneumonia and wound infection is that 41 patients in our series (42%) presented with intestinal obstruction secondary to cancer, a predisposing factor for aspiration pneumonia. Thirtyfive (36%) patients (17 cancer, 18 complicated diverticulitis) presented with large bowel perforation complicated, resulting in intra-abdominal soilage, could account for our rate of wound infection. In our series, the prevalence of diabetes (20%) and respiratory problems (10%) predisposed towards developing wound infection and pneumonia. While the correlation between operative time and mortality or morbidity was not statistically significant, a longer operative time showed a trend towards an increase in morbidity (p=0.08). We found the mortality and morbidity rates not influenced by the aetiology or surgeon’s expertise. A possible explanation for this is the fact that consultant surgeons were involved in the more difficult and complex
cases with higher risks involved, whereas the more straightforward cases were performed by the registrars. Our reversal rate of 40% is lower than those reported in most series [3, 4, 6]. We believe that this is due to the fact that the predominant indication for Hartmann’s procedure in our institution was colorectal cancer (59%) which resulted in death before reversal could be carried out. This is supported by our analysis which revealed a trend towards benign diseases being reversed. The largest series in literature by Desai et al. [6], comprising 185 patients, had a reversal rate of 57%. This series comprised a comparably larger proportion of patients with complicated diverticulitis (58%) and reflects the situation in the west. Their reversal rate for cancer was only 13% compared to 70% for diverticulitis. In another series by Zorcolo et al. [10], 39.2% of 102 patients had their stomas reversed and the reversal rate for cancer was 8.7% compared to 48.1% for diverticulitis. Our reversal rates for colorectal cancer and complicated diverticular disease were 30% and 57%, respectively. The reasons for not reversing the stoma in the remaining 42 patients include high risk due to comorbidities in 17 patients (40%). This is consistent with the fact that 15 patients (36%) had an ASA score of 3 or 4 when Hartmann’s procedure was performed. Comorbidities increase with age and the risks of the reversal procedure can be based on the patient’s comorbidities. In our series, increased age was the only factor significantly related to a lower likelihood of stoma reversal. Five patients (12%) refused the reversal procedure when it was offered. The reasons given included fear of undergoing another major operation and finding the stoma a convenient form of toilet in 2 patients (5%). These patients were bedridden preoperatively and an end colostomy made nursing easier. Hence, our permanent stoma rates were more a reflection of the disease and patient profile, and not our surgical expertise. In the Asian population, where colorectal cancer carries a high incidence, Hartmann’s procedure is a valid option and has an important role in patients who present as an emergency case. Its mortality rate is not higher than in other parts of the world although it carries a significant morbidity. This is associated with the fact that most patients are elderly, with pre-existing medical problems. The event of non-reversal was related to the underlying disease and patient profile and is often performed as a definitive procedure.
References 1. Hartmann H (1921) Chirurgie du rectum. Masson, Paris 2. Boyden AM (1950) The surgical treatment of diverticulitis of the colon. Ann Surg. 132:94–109
Tech Coloproctol (2008) 12:21–25 3. Khosraviani K, Campbell WJ, Parks TG, Irwin ST (2000) Hartmann procedure revisited. Eur J Surg 166:878–881 4. Schein M, Decker G (1998) The Hartmann procedure: extended indications in severe intraabdominal infection. Dis Colon Rectum 31:126–129 5. Seah DW, Ibrahim S, Tay KH (2005) Hartmann procedure: Is it still relevant today? ANZ J Surg 75:436–440 6. Desai DC, Brennan EJ, Reilly JF, Smink RD Jr (1998) The utility of the Hartmann procedure. Am J Surg 175:152–155 7. Banerjee S, Leather AJM, Rennie JA et al (2005) Feasibility and morbidity of reversal of Hartmann’s. Colorectal Dis 7:454–459 8. Salem L, Flum DR (2004) Primary anastomosis or Hartmann’s procedure for patients with diverticular peritonitis? A systemic review. Dis Colon Rectum 47:1953–1964 9. Koruth NM, Krukowski ZH, Youngson GG et al (1985) Intraoperative colonic irrigation in the management of left-sided large bowel emergencies. Br J Surg 72:708–711 10. Zorcolo L, Covotta L, Carlomagno N, Bartolo DCC (2003)
25
11.
12. 13.
14.
15. 16.
Safety of primary anastomosis in emergency colo-rectal surgery. Colorectal Dis 5:262–269 Kressner U, Antonsson J, Ejerblad S et al (1994) Intra-operative colonic lavage and primary anastomosis - an alternative to Hartmann procedure in emergency surgery of the left colon. Eur J Surg 160:287–292 Lee YS (1986) Diverticular disease of the large bowel in Singapore: an autopsy study. Dis Colon Rectum 29:330–335 Chia JG, Wilde CC, Ngoi SS et al (1991) Trends of diverticular disease of the large bowel in a newly developed country. Dis Colon Rectum 34:498–501 Biondo S, Jaurrieta E, Marti Rague J et al (2000) Role of resection and primary anastomosis of the left colon in the presence of peritonitis. Br J Surg 87:1580–1584 Singapore Department of Statistics Year 2000 Huang J, Seow A, Shi CY, Lee HP (1999) Colorectal carcinoma among ethnic Chinese in Singapore: trends in incidence rate by anatomical subsite from 1968 to 1992. Cancer 85:2519–2525
