| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2006;29:244-247
© 2006 Elsevier Science NL
Department of Thoracic Surgery, Nottingham City Hospital, Hucknall Road, Nottingham NG5 1PB, United Kingdom
Received 21 April 2005; received in revised form 6 November 2005; accepted 8 November 2005.
* Corresponding author. Tel.: +44 7834600435. (Email: edblackis{at}hotmail.com).
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Methods3. Results4. ConclusionsReferences |
|---|
Key Words: Oesophageal cancer • Oesophageal surgery • Splenectomy
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. ConclusionsReferences |
|---|
|
2. Methods |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. ConclusionsReferences |
|---|
Operations were performed through a left-sided approach for mid-lower third tumours and a right-sided two-stage approach for the rest. Occasionally, a cervical anastamosis was performed, otherwise a stapled intra-thoracic anastamosis was preferred. If the spleen was injured during an operation, packing for several minutes was usually the first manoeuvre. If this did not stop the bleeding, splenectomy rather than splenic preservation, by several of the available techniques, was carried out. Postoperative care was standardised for all patients which included nursing them on a high-dependency ward and keeping them nil by mouth until water-soluble contrast swallow on the seventh postoperative day. Rigorous attempts were made to prevent atalectasis with twice-daily physiotherapy, regular nebulised medication and judicious use of mini-tracheostomies.
In the event of splenectomy, a standardised regime of prophylaxis was used as described in the protocol from our Department of Microbiology. Our regime was pneumococcal vaccine (Pneumovax® II) repeated every 3–5 years.
When it became available, a one-off Haemophilus influenzae type b (Hib) and meningococcal Group C conjugate vaccine was given to all who had not been previously immunised. A recommendation was sent to all primary care doctors to administer yearly vaccination with meningococcal polysaccharide A+C vaccine.
2.1 Prophylactic antibiotics
After finishing intravenous antibiotics with benzyl penicillin all patients received antibiotic prophylaxis for at least 2 years following splenectomy with Penicillin V 250 mg bd (or erythromycin 250 mg bd).
To help with compliance problems, patients were educated and followed up. In the event of infection their primary care physicians were recommended to administer amoxycillin or erythromycin.
Variables examined as possible risk factors for splenectomy were age, sex, neoadjuvant chemotherapy, body mass index, resection margins, tumour length, surgical approach, site of tumour, type of tumour and pathological stage. Outcome variables examined were postoperative blood use, pulmonary complications, cardiac complications, anastomotic leaks, hospital deaths, survival and wound infections. Variables were analysed to determine the association with inadvertent splenectomy and their likely impact on in-hospital mortality and long-term survival. Continuous variables were analysed by Student's t-test or Mann–Whitney and categorical variables by chi-squared analysis. Those variables that had a p value of less than or equal to 0.1 were put into a logistic regression model to determine significance. Long-term survival was assessed by Kaplan–Meier method, with log-rank test for differences and Cox model to assess the impact of variables that approached significance.
2.2 Definitions
In-hospital death was defined as death within the same hospital admission or within 30 days.
Pulmonary complications were evidence of either segmental collapse/infection/effusion/requirement for ventilatory assistance.
Leaks were defined as clinical or radiological evidence of mediastinal leak.
Wound infection was defined as a wound requiring intravenous antibiotics and/or debridement.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. ConclusionsReferences |
|---|
|
|
|
|
4. Conclusions |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. ConclusionsReferences |
|---|
It has been previously reported that the cause of iatrogenic trauma to the spleen is most often traction on splenic ligaments or adhesions [9,10]. We had a very significant difference in incidence of splenectomy between the different types of operations used at our institution (Table 1). We believe that the reasons for this are multiple but might include the direct access to the spleen and protection from traction pressure from retractors in the thoracophrenotomy operation. During thoracolaparotomy, the retractor tends to press on the spleen. The Ivor–Lewis operation has the worst access to the spleen. In an obese patient, particularly when combined with short stature, access to the fundus and greater curve of the stomach is difficult. The peritoneal–splenic adhesions can also be missed. In our hands, the thoracophrenotomy approach was the least likely to be associated with splenic injury. Also, we found more female patients who underwent splenectomy (40% vs 27%; p = 0.06). This may have been also because of worse access to the fundus from their body habitus and shorter short gastrics.
There were no differences in tumour cell type but there was an increase in the incidence of splenectomy in the higher N stage disease. It is also usual that once committed to resection, upon encountering bulky N2 area nodal disease, we tend to continue with resection rather than abandon at this stage. We were a little surprised to find no association with the advanced T stages. It has also been our experience that in patients who have more intra-abdominal fat, or during an Ivor–Lewis oesophagectomy, it can be difficult to get access to the vasa brevia. This area is easy to access during the left-sided approaches, which centre over the fundus and splenic hilum. We think that this accounts for the almost significant increase in splenectomy in women and Ivor–Lewis approaches (Table 1).
Data have been available for a long time about the association between splenectomy and an increased risk of infections, both in paediatric and adult populations [1,3]. There might also be an increased risk of non-infective complications post-splenectomy [2]. Furthermore, patients with underlying diseases, such as cancer, may be at a greater risk than patients who had a splenectomy from trauma [3]. It seems that patients who have inadvertent splenectomy during surgery for intra-abdominal malignancy have a worse outcome than those who have splenectomy for trauma [4].
Sepsis following splenectomy is rare [4]. However, we know that following D2 gastrectomy, mortality is increased significantly by the additional insult of splenectomy [10,11]. Following oesophagectomy specifically, Kyriazanos et al. [5] found a significant increase in many infective complications. We had a more varied picture. Whilst we had no increase in pulmonary complications or wound infections we had a significant increase in leaks (Table 2). We could not identify a cause for this; for example, there was not a difference in cervical anastamotic rate. Perhaps this difference was due to a generally more difficult case, for example, resulting in non-specific trauma to both conduit and spleen during access.
There is no consensus about how to manage the patient who has an inadvertent splenectomy, in part, due to studies highlighting the relatively low incidence of infective complications [4]. We treat our patients aggressively as guided by consultation with our microbiologists. They also provide a protocol for the management of all patients who have splenectomy. It consists of antibiotic, immunological and educational prophylaxis as detailed earlier in the text. We are not aware of any complications from this regime. Adverse reactions to the vaccines were not seen. Not surprisingly, the patients who had splenectomy had greater need for blood transfusions postoperatively (p = 0.03) (Table 2). The mean number of units transfused was still only less than 2 units.
In the absence of other data on infective complications, we assessed pulmonary complications, wound infections and SVT. There were no significant differences between the groups. Similar to the findings of Kyriazanos et al. [5], we had more leaks in the splenectomy group, though the rate was still only 14.7% in the splenectomy group (p = 0.02). Our microbiologists recommend post-splenectomy immunoprophylaxis against pneumococcus, meningococcus and H. influenzea, chemoprophylaxis, intensive postoperative care and patient education to protect patients. Perhaps our routine use of these strategies helped to control infections. A prospective trial would be required to know for sure.
We examined the early (30-day) complications as well as studied the late survival. Perioperative mortality was not increased by splenectomy, despite the increase in the number of leaks. Kyriazanos et al. [5] reported increased hospital mortality from 8 to 36% (p < 0.01). The cause seemed to be a significant increase by more than 25% in the incidence of pulmonary complications (p < 0.01). In the light of the increased mortality attributable to splenectomy in patients randomised to D2 gastrectomies, our findings are encouraging [10,11]. That our mortality rate remains quite low in both groups may be due in part to the fact that our unit is a high volume centre for oesophagectomies and experience is thus translated into better outcomes [12].
Median survival was not significantly different either at 627 (554–700) or 551 (332–770) days for no-splenectomy and splenectomy groups, respectively (p = 0.63). Most of our patients had stage pT3N1 adenocarcinoma. Davis et al. [13] found a reduced 5-year survival in patients undergoing colectomy for cancer if they also had a splenectomy. While there was no difference in 30-day mortality, the 5-year survival was 44% with splenectomy compared to 62% without splenectomy (p < 0.03). Conversely, the impact of splenectomy in D2 gastrectomy has been reported not to affect 5-year survival [11].
In conclusion, splenic injury during oesophagectomy is an infrequent occurrence with little available literature on the significance of its impact on outcome. In our series, which represents a large single institutional series, splenectomy was least likely with those with the left thoracophrenotomy approach and early N stage and possibly with males. It seems to be associated with a greater tendency for leaks. In our institution, we have an aggressive and standardised approach to infection control. Perhaps this is an important reason that we did not experience greater infections in this group.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstract1. Introduction2. Methods3. Results4. ConclusionsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
