Paraesophageal hernias typically occur in older patients, Open Repair of Paraesophageal Hernia: Reassessment of Subjective and Objective Outcomes

Transcription

1 Open Repair of Paraesophageal Hernia: Reassessment of Subjective and Objective Outcomes Donald E. Low, MD, FACS, and Trisha Unger, MD Section of General Thoracic Surgery, Virginia Mason Medical Center, Seattle, Washington Background. Surgical repair of paraesophageal hernias (PEH) remains a challenging operation. Increasing numbers of patients are undergoing laparoscopic repair. This series provides an up-to-date benchmark of the results of open repair of PEH to compare with current laparoscopic series. Methods. All patients undergoing surgical repair of PEH by a single surgeon between April 1996 and November 2001 were included. Follow-up included postoperative SF-36 survey and objective reassessment (barium swallow or endoscopy) at a mean of 29.8 months. Results. Seventy-two consecutive patients (mean age, 68.7 years) presented with large PEH. Presenting symptoms included heartburn (60%), dysphagia (43%), chest pain (42%), anemia (39%), and dyspnea (32%). Surgical treatment involved transabdominal open repair including sac excision, crural closure (primary closure 98.6%), and antireflux procedure (Hill procedure, 96%; Nissen fundoplication, 4%). No patient required a Collis procedure. Postoperative assessment (subjective, 97%; objective, 88%) was prospective. Median operative length was 155 minutes. Median length of stay was 4.5 days (range, 3 to 12 days). Postoperative complications occurred in 17 of 72 patients (23.6%), but no patient sustained intraoperative or postoperative visceral injuries. In-hospital and 30-day mortality was zero. Heartburn and dysphagia symptom scores demonstrated significant improvement (p < 0.001). Postoperative SF-36 scores demonstrated levels better than the general population (six of eight categories) and better than the age-matched population (eight of eight categories). Objective follow-up demonstrated recurrent hernias in 11 patients (18%). Most of these recurrences (73%) were less than 2-cm sliding hernias, and no patients required revisional surgery. Conclusions. The results of open repair of PEH have continued to evolve in the same time that has seen the introduction of laparoscopic PEH repairs. Results with the open approach in the modern era can provide excellent outcomes, which are comparable to and in some measures exceed those obtained with the laparoscopic approach. (Ann Thorac Surg 2005;80:287 94) 2005 by The Society of Thoracic Surgeons Accepted for publication Feb 1, Address reprint requests to Dr Low, Virginia Mason Medical Center, Section of General Thoracic Surgery, 1100 Ninth Ave, Seattle, WA 98111; gtsdel@vmmc.org. Paraesophageal hernias typically occur in older patients, frequently associated with other comorbid medical issues. Although seen much less commonly than standard sliding (type 1) hiatal hernias with the aging of the population, the incidence and recognition of paraesophageal hernias, previously thought to be less than 0.5% of all hiatal hernias [1], is likely increasing. Paraesophageal hernias have three different presentations. Type 2 or true paraesophageal hernias are the most uncommon. These hernias have no sliding component and have the esophagogastric junction located in its normal anatomic location, below the diaphragm. The most common configuration is the mixed type 3 paraesophageal hernia, which presents with both a sliding and paraesophageal component. Type 4 hernias can be either type 2 or type 3 in orientation, but also include other viscera in the hernia sac, typically transverse colon or small bowel. Surgeons have typically advocated elective repair of these hernias because of the perceived risk of serious complications, particularly incarceration and vascular compromise of the stomach [2, 3]. There is general agreement among surgeons that paraesophageal hernia repair is a challenging procedure, with higher morbidity and mortality rates compared with other elective esophageal and gastric procedures. Outcomes assessment is increasingly providing information beyond morbidity and mortality, specifically quality-oflife factors, cost assessments, and recurrence rates, which should be the criteria used to decide when and how these procedures are performed. Recently there have multiple reports documenting the results of laparoscopic paraesophageal hernia repair. The laparoscopic approach has been increasingly used throughout the last decade because of a perceived advantage of minimizing the impact and potentially the incidence of complications. There is currently no series of open procedures done exclusively during the same period that has seen the introduction of laparoscopic approach. A reappraisal of the results of the open repair applying up-to-date methods of preoperative assessment, operative technique, and pain management pro by The Society of Thoracic Surgeons /05/$30.00 Published by Elsevier Inc doi: /j.athoracsur

2 288 LOW AND UNGER Ann Thorac Surg OPEN REPAIR OF PARAESOPHAGEAL HERNIA 2005;80: Table 1. Open Paraesophageal Hernia Repair: Outcomes Table 3. Paraesophageal Hernia Characteristics Variable Mean Median Range PEH Size a PEH Type Operative length (min) Postoperative NG (days) Hospital stay (days) NG nasogastric tube. vides a more appropriate comparison with current laparoscopic series. Patients and Methods Patients Between April 1996 and November 2001, 72 consecutive patients underwent surgical repair of paraesophageal hernia by a single surgeon at Virginia Mason Medical Center. Mean age was 68.7 years (range, 45 to 91 years), with a male-to-female ratio of 31:41. Preoperative assessment included barium swallow, endoscopy, and esophageal manometry in all patients. Twenty-four hour ph studies were done selectively. All patients had type 2, type 3, or type 4 paraesophageal hernias assessed on preoperative barium study, and patients were not included unless the hernia involved at least 50% of the stomach. Preoperative, intraoperative, and postoperative data were gathered prospectively. Standard follow-up involved assessment in the surgical clinic at 1 week, 3 months, and 6 months. Preoperatively and after discharge patients completed antireflux symptom surveys developed by Pope [4]. All patients were reassessed at a mean of 29.8 months after surgery (range, 2 to 84 months) to undergo standardized Pope questionnaire, SF-36 quality-of-life survey, and objective reassessment with barium swallow or upper endoscopy. Sixty patients (83%) had follow-up of at least 1 year. All barium studies were independently reviewed by a single radiologist with a specific interest in gastrointestinal radiology. Statistical analysis was performed with Student s paired t test. This study was reviewed and accepted by our institutional review board. Barium studies performed without clinical indications were preceded by informed consent and reimbursed from the study budget. Table 2. Paraesophageal Hernia Repair: Presenting Symptoms Heartburn 43 60% Dysphagia 31 43% Chest pain 30 42% Regurgitation 13 18% Dyspnea 16 22% Anemia 28 39% UGI bleeding 9 13% Bloating 8 11% Weight loss 8 11% 50% 25 II % 33 III % 14 IV 18 colon: 12 small bowel: 6 a Denotes percentage of stomach above the diaphragm at time of preoperative barium study. PEH paraesophageal hernia. Surgical Technique All repairs were transabdominal through an upper midline incision with patient-controlled epidural anesthesia catheters placed preoperatively. Operative approach involved removal of the paraesophageal hernial sac in all patients. Diaphragmatic closure was accomplished primarily in 71 patients (98.6%). One patient required reinforcement with Surgisis (Cook Surgical, Bloomington, IN). An antireflux procedure was done on all patients, predominantly the Hill operation in 69 patients (96%); the remainder underwent Nissen fundoplications, 3 patients (4%). No Collis procedures were required. The Hill procedure has been described in detail elsewhere [5, 6]. It is unique among the established antireflux procedures in that it is predicated on firm attachment of the anterior and posterior phrenoesophageal ligaments to reliable posterior attachments such as the condensation of the crus. This process reestablishes the gastroesophageal flap valve to control reflux, but also firmly anchors the esophagogastric junction within the abdomen, decreasing the chances of recurrent hiatal hernia. It is this process of reliably anchoring the repair that makes the Hill operation ideally suited for treatment of paraesophageal hernias. Table 4. Paraesophageal Hernia Repair: Morbidity and Mortality Initial Delayed Atrial fibrillation 6 Heartburn 7 (10%) Pulmonary 6 Dysphagia GI ileus 3 Transient 6 (9%) GU Required dilation 5 Retention 1 Gas bloat 2 Infection 2 Incisional hernia 6 Delirium 1 Early or late reoperations 0 Epidural headache 1 Angioedema 1 Visceral injury or 0 postoperative leak Mortality In-hospital or 30-day 0 UGI upper gastrointestinal. GI gastrointestinal; GU genitourinary.

3 Ann Thorac Surg LOW AND UNGER 2005;80: OPEN REPAIR OF PARAESOPHAGEAL HERNIA 289 Table 5. Paraesophageal Hernia Repair: Postoperative Symptom Assessment a Variable Symptom Preoperative Postoperative p Value Intensity Heartburn Dysphagia Frequency Heartburn Dysphagia a From Pope CE. World J Surg 1992;16: Additional gastric fixation was carried out in 28 patients (39%), gastropexy in 20 patients, gastrostomy in 8 patients. Gastropexy or gastrostomy placement was typically used in elderly or immunosuppressed patients with very large hiatal defects. Gastrostomy tubes were also inserted in 4 patients with previous extensive abdominal surgery and history of ileus. Two procedures (3%) were urgent. Three (4%) were redo operations. Operative details are outlined in Table 1. Median operative length was 155 minutes, and median hospital stay was 4.5 days. Results Presenting symptoms or related issues such as anemia and weight loss were present in all patients. Presenting symptoms included heartburn (60%), dysphagia (43%), chest pain (42%), anemia (39%), and dyspnea (32%; Table 2). Comorbidities were common in this elderly population, with 42% of patients demonstrating American Society of Anesthesiologists classification 3 or 4. Hernia size and type is shown in Table 3. Postoperative complications are reviewed in Table 4. Preoperative motility was normal in 66 patients (92%). The other 6 patients were found to have nonspecific motility disorders. Only 5 patients required postoperative dilation for dysphagia, and 2 of these patients (40%) had preoperative nonspecific motility disorders. Overall incidence of postoperative complications was 17 of 72 (23.6%). No patient had intraoperative or postoperative visceral injuries or leaks, and in-hospital and 30-day mortality was zero. Immediate postoperative symptom assessment demonstrated statistically improved intensity and frequency levels of heartburn and dysphagia (Table 5). Delayed complications included predominantly heartburn (10%), dysphagia (9%), and incisional hernia (9%) (Table 4). No patient has required revisional surgery for recurrent hiatal hernia. Mean long-term follow-up was 29.8 months. Four patients died in the postoperative period secondary to issues unrelated to their paraesophageal hernia repair, leaving 68 available for subjective and objective assessment. Two patients were lost to follow-up, resulting in subjective follow-up being available in 66 of 68 patients (97%). Four additional patients were in nursing homes, and 2 refused objective reassessment, resulting in 60 of 68 patients (88%) undergoing either repeat barium study (87%) or EGD (13%) postoperatively. Postoperative SF-36 assessments were carried out in 97% of patients. Table 6 demonstrates that mean SF-36 scores exceeded national means in the general population in six of eight factors and exceeded all factors in the age-matched population. In addition, all but 1 patient indicated they would choose to have the operation again (98.6%). Objective assessment was carried out by nonsurgical physicians, and results are shown in Table 7. Normal studies were demonstrated in 49 patients (82%). Recurrent sliding hernias were seen in 16.6%, although 8 of 10 of these were measured between 1 and 2 cm. No recurrent hernia was believed to exceed 4 cm, and only 1 patient had a 3-cm recurrent paraesophageal hernia. We did not think that the single patient with a recurrent paraesophageal hernia required operative revision because she was 80 years old, had previously had a hernia involving 100% of her stomach, and postoperatively had no symptoms. Comment Minimizing morbidity and mortality while demonstrating real improvements in symptoms, quality of life, and durability of results remain the hallmarks of any surgical treatment. Laparoscopic surgery has had significant impact on all of these factors in a variety of surgical procedures, particularly cholecystectomy and operations for gastroesophageal reflux disease. Discussion continues regarding the ultimate role of laparoscopic surgery in other areas such as appendectomy, inguinal hernia, and colon resection for cancer. Multiple recent publications have demonstrated the Table 6. Postoperative Quality-of-Life Assessment SF-36 Factors Current Series Mean National Mean a National Age-Matched Mean a Health perception Physical function Physical health Emotional health Social functioning Mental health Bodily pain Energy vitality a From Ware JE, Snow KK, Kosinski M. SF-36 health survey manual and interpretation guide. Boston: The Health Institute New England Medical Center, 1993.

4 290 LOW AND UNGER Ann Thorac Surg OPEN REPAIR OF PARAESOPHAGEAL HERNIA 2005;80: Table 7. Paraesophageal Hernia Repair: Recurrence Rates a Normal study 49 (82%) Recurrent type 1 hernia (16.6%) 1 2 cm cm 2 4 cm 0 Recurrent PEH 1 (1.6%) a n 60 patients (88%); mean follow-up was 30 months. Objective follow-up was obtained using upper gastrointestinal barium studies in 52 patients and upper endoscopy in 8 patients. All postoperative barium studies were reviewed by a single radiologist with special interest in gastrointestinal radiology. PEH paraesophageal hernia. technical feasibility of laparoscopic paraesophageal hernia repair [7 15]. If there is a theme that is typical in these reports, it is that these repairs are much more challenging than standard antireflux procedures and typically are associated with a higher incidence of morbidity and mortality and a defined learning curve. Before the mid-1990s the majority of paraesophageal hernia repairs were open procedures, either transabdominal or transthoracic. There are multiple publications [1, 16 22] demonstrating levels of morbidity and mortality, which until recently have been the standard of comparison for other surgical series (Table 8). These series of open repairs typically involve sizable numbers of patients (n 37 to 124), but have involved study periods ranging from 9 to 36 years. Although symptomatic results are routinely good, specific quality-of-life assessments have not been routinely included. In addition, assessment of technical and recovery issues such as operative length and length of hospital stay were reported irregularly, and even series reporting results in the 1990s show a mean postoperative length of stay of more than 9 days [1, 21, 22]. Visceral injuries are reported in 1% to 4% of cases, and perioperative mortality ranges from 0% to 2%. Although the majority of laparoscopic paraesophageal hernia reports indicate overall good outcomes, enthusiasm for the laparoscopic approach has not been universal. Dahlberg and colleagues [23] reported the Mayo Clinic results in 37 patients. Good to excellent symptomatic results were demonstrated in 83.9%. However, owing to an esophageal leak rate of 5.4%, reoperative rate of 8%, and a mortality rate of 5.4%, they summarized that laparoscopic paraesophageal hernia repair could not currently be recommended as the standard approach. There are currently no randomized comparisons of laparoscopic versus open paraesophageal hernia repairs. Schauer and associates [24] have reported a nonrandomized comparison of laparoscopic versus open repair in which all the laparoscopic repairs were performed by the two lead authors, whereas the open operations were performed by a larger group of affiliated surgeons with variable degrees of experience in paraesophageal hernia repair. Mean hospital stay was 10.3 days in the open group and 4.98 days in the laparoscopic group. The only patients requiring reoperation were in the open group, 2 (8%), although follow-up was much longer in the open operations, 48 months, versus the laparoscopic repair, 13 months. Although the overall incidence of complication was higher in the open group, 3 patients in the laparoscopic series experienced delayed esophageal leaks requiring reoperation, and the only death occurred in the laparoscopic group. A report by Hashemi and colleagues [25] compared outcomes and recurrence rates in a nonrandomized series of open and laparoscopic paraesophageal hernia repairs. They report that good and excellent outcomes favored the patients having open procedures (88%) versus those having laparoscopic operations (76%). It was also noted that in spite of the fact that the follow-up was longer in the open versus laparoscopic groups (35 versus 17 months), recurrent hiatal hernias were much more commonly seen after laparoscopic versus open procedures (42% versus 15%). The complexity of laparoscopic repairs of paraesophageal hernias is highlighted in a paper by Trus and coworkers [26], specifically reviewing the complications associated with 76 patients at two major laparoscopic institutions. Visceral injuries occurred in 11% of patients intraoperatively and 4% postoperatively. Reoperations were required in 9%, and mortality rate was 3%. The authors highlight that their results improved with experience, but that paraesophageal hernia provides chal- Table 8. Open Repairs of Paraesophageal Hernia Reference (n) Study Period (y) Excellant or Good Results Operation Length (min) Length of Stay (days) Visceral Injuries Reoperation Mortality Hernia Recurrence Allen et al 1993 [1] % % 1% 1% 2% Altorki et al 1998 [18] % N/A N/A 2% 0 2% N/A Ellis et al 1986 [21] % N/A 9.5 N/A 7% 2% N/A Geha et al 2000 [17] % N/A N/A N/A 2% 2% 2% Martin et al 1997 [22] % N/A 9.1 N/A 4% 0 N/A Maziak et al 1998 [20] % N/A N/A 4% 4% 2% 2% Myers et al 1995 [19] % N/A N/A 2.7% 2.7% 0 N/A Williamson et al 1993 [16] % N/A N/A 1% 7% 1.7% 11% N/A not available.

5 Ann Thorac Surg LOW AND UNGER 2005;80: OPEN REPAIR OF PARAESOPHAGEAL HERNIA 291 Table 9. Laparoscopic Repairs of Paraesophageal Hernia Reference (n) Study Period (y) Operation Length (min) Length of Stay (days) Visceral Injuries Conversion Reoperation Mortality Hernia Recurrence Dahlberg et al 2001 [23] % 5% 8% 5.4% 14% (22 patients) Diaz et al 2003 [7] % 2.5% 3% 1.7% 32% (80 patients) Gantert et al 1998 [10] % 6% 5% 2% N/A Horgan et al 1999 [14] % 5% 0 2% N/A Jobe et al 2002 [9] % 8% 4% 0 32% (34 patients) Mattar et al 2002 [12] % 2% 2% 2% 33% (32 patients) Perdikis et al 1997 [13] 65 5 N/A 2.0 5% 3% 5% 0 N/A Pierre et al 2002 [8] % 1.5% 2.5%.5% N/A Swanstrom et al 1999 [15] % 0 4% 0 N/A Trus et al 1997 [26] % 1.5% 8% 3% N/A Wiechmann et al 2001 [11] % 10% 7% 2% N/A N/A not available. lenges and potential risks that far exceed standard operations for gastroesophageal reflux disease. They also indicate that even these early results were comparable to open series of paraesophageal hernia repairs available at the time. This statement is accurate to the extent that satisfaction and mortality rates were comparable. However, the incidence of intraoperative and postoperative visceral injuries (greater than 10%) and reoperation rates (9%) exceeds those reported in virtually any open series (Table 8). Table 9 demonstrates that the current laparoscopic paraesophageal hernia series have involved assessment of significant numbers of patients (n 37 to 203), but have accumulated their experience in a much faster and more current time frame (4 to 9 years) than reports of open series. The increased complexity of these operations is demonstrated in reported operative lengths varying between 3 and 4 hours, with length of hospital stays varying between 2.0 and 4.2 days. These figures are significantly higher than results reported in standard antireflux operations. When compared with previously published series of open repairs with respect to hospital length of stay, even the most recent publication still indicated an average length of stay of 9 days (Table 8). Our series has shown a median hospital stay of 4.5 days and a mean stay of 4.76 days, not profoundly different than many of the laparoscopic reports. In addition, the only previous open series to report operative times is by Allen and associates [1], which in elective repairs was a median of 150 minutes, very comparable to our series (median, 155 minutes; mean, 162 minutes) and significantly shorter than the majority of laparoscopic operations. Further proof of the complexity of the laparoscopic operations highlighted in Table 9 is the incidence of conversion in laparoscopic operations (0% to 10%), reoperation rates (0% to 8%), and particularly the incidence of intraoperative and postoperative visceral injuries (1% to 12%). Many of these reports are from some of the most accomplished laparoscopic centers, supporting the concept that laparoscopic paraesophageal hernia repair should not be done except in highly experienced units. Objective follow-up has not routinely been a component of postoperative assessment, and when it has been reported, it has only occasionally involved an attempt to restudy all patients. Maziak and coworkers [20] restudied all 94 of the patients who underwent open transthoracic repairs. Unfortunately, the actual incidence and dimensions of recurrent hernias are not reported. Hashemi and colleagues [25] reported recurrent hernias in 42% of laparoscopic repairs and 15% of open operations, but noted that hernia recurrence did not have a direct relationship with the presence or absence of recurrent symptoms. Objective follow-up was accomplished in 88% of the current series. Of the 11 recurrences discovered, 8 (73%) were 2 cm or less in greatest dimensions. This results in an overall recurrence rate of 18% when including hernias of any size, 5% in recurrent hernias greater than 2 cm, and 1.6% for recurrent paraesophageal hernias. The clinical significance of these recurrent hernias is currently not known. Several reports [7, 9, 25] have suggested that there is no discernible association with the recurrent hernias and short-term symptom recurrence. It is very unlikely that the small recurrent sliding hernias will be clinically relevant in this elderly population. However, recurrent large sliding or paraesophageal hernias are a common reason for reoperation in both open and laparoscopic series [7, 9, 11, 16, 21, 23, 26]. There is also a trend toward higher recurrence rates after operations that did not include an antireflux operation [16, 21, 27]. It would appear that the incidence of recurrent hernias after paraesophageal hernia repair is significantly higher than rates published in reports on patients having operations for uncomplicated gastroesophageal reflux disease. This series documents objective follow-up in a very high percentage of patients and provides a benchmark for comparison with respect to present and future open and laparoscopic series. One of the most plausible explanations for higher

6 292 LOW AND UNGER Ann Thorac Surg OPEN REPAIR OF PARAESOPHAGEAL HERNIA 2005;80: recurrence rates associated with paraesophageal hernia repair would be the potential influence of a short esophagus. The actual incidence of shortened esophagus and paraesophageal hernia is unknown, although it has been variously reported between 0% and 80% [9, 10, 20, 28, 29]. Trus and colleagues [26] suggested that esophageal foreshortening is likely a contributing factor associated with difficult hernia reduction and hernia recurrence. Multiple additional reports have indicated the importance of adequate esophageal mobilization and in selected cases for esophageal lengthening or Collis procedures to limit postoperative recurrences [14, 18, 27]. However, in both open and laparoscopic series there remain differences of opinion regarding the actual number of patients who require a Collis procedure at the time of their paraesophageal hernia repair. The Collis procedure is seeing increasing application in laparoscopic series, best illustrated by the reports from the University of Pittsburgh, which currently has the single largest published experience with laparoscopic paraesophageal hernia repair. In their first report, by Luketich and associates [29], 100 patients underwent laparoscopic repair of paraesophageal hernia with 27% having associated Collis operations. In a follow-up report from the Pittsburgh group by Pierre and coworkers [8], 203 patients had laparoscopic repairs and 55% had Collis operations. Assuming that the second report included patients from their initial series, this would imply that this very experienced group has evolved to using Collis operations in the majority of their procedures. This occurred in spite of the fact that outcomes in their first series were excellent (satisfaction score, 91%). There is currently only one assessment of results in patients who have undergone laparoscopic Collis gastroplasty for short esophagus. Jobe and colleagues [30] provided objective follow-up of 14 of 15 patients after laparoscopic Collis procedures. They demonstrated that 50% of patients had abnormal postoperative 24-hour ph scores (mean DeMeester score of 100), 36% had esophagitis, and 43% had distal aperistalsis in the neoesophagus. These results suggest that the Collis procedure, at least when applied laparoscopically, requires continued reassessment as the technical approach to the laparoscopic Collis operation continues to evolve. The importance of adequate esophageal mobilization should not be underestimated. The fact that esophageal mobilization can be accomplished more effectively through an open thoracotomy secondary to better access to the thoracic esophagus or laparoscopically secondary to better visualization has some proponents. We have found that the esophagus can be extensively mobilized transabdominally through the dilated hiatus after removal of the hernial sac. We have also found it unnecessary to carry out a lengthening procedure because of our tendency to use the Hill operation, which is the single antireflux procedure predicated on anchoring the repair to firm intraabdominal structures such as the preaortic fascia or the condensation of the crus. It is for this reason that we believe that the Hill operation is particularly well suited for paraesophageal hernia repair. Virtually all surgeons would agree that paraesophageal hernia repair is a technically challenging operation typically done in elderly patients, some who will present with minimal or no apparent symptoms. Publications from the 1960s through the 1980s have indicated significant risk of incarceration and strangulation, with mortality rates ranging from 17% to 50% [2, 3, 28]. A recent meta-analysis has suggested that the incidence of true surgical emergencies has been overestimated. The same report demonstrates that 18% of 65-year-old patients will develop acute symptoms. In spite of these findings, Stylopoulos and coworkers [31] conclude that many of these patients can be managed with watchful waiting. It is our contention that the incidence of symptoms associated with paraesophageal hernias is underappreciated, and the only study to prospectively assess symptom changes with time, by Treacy and associates [32], demonstrated progression of symptoms in 45% of patients who were not managed surgically. Much more germane to the clinical decision-making process regarding paraesophageal hernias has been the impression that a significant percentage of these patients are asymptomatic. It is our belief that few of these patients do not have some clinical or symptomatic issues. These hernias enlarge slowly with time and can reach massive proportions. More subtle issues such as pain or bloating with eating, weight loss, or anemia are often relegated to the fact that these patients are typically elderly. In addition, we have recently documented that these hernias can have a significant affect on breathing mechanics [33]. Statistically significant improvements in spirometry values (ranging from 16% to 20% improvements in forced expiratory volume in 1 second and forced vital capacity) can be expected in patients having paraesophageal hernia repairs. Two patients in this series who were on home oxygen were able to discontinue their oxygen postoperatively. These findings would indicate that patients who were previously believed to be poor surgical candidates because of their breathing problems may now have the most compelling reasons to seek elective repair. In the current series, all patients were believed to have either symptoms or clinical issues related to their large paraesophageal hernias. Similar to other studies, we have shown that these patients will show improvement with respect to standard symptoms of heartburn and dysphagia as well as comparing favorably to the general and age-matched populations with respect to postoperative quality-of-life scores (Table 6). It has been assumed that laparoscopic repairs are superior to open operations because of less impact on patients, potentially shorter hospital stays, and lower morbidity and mortality. Streets and colleagues [34] have shown that with respect to quality-of-life measurements after standard antireflux operations, it was not the invasiveness of the approach that ultimately dictates outcome. Comparisons of Tables 8 and 9 will show that mortality and reoperation rates are similar, but tend to favor the open approach. Historic measurements of length of stay clearly favor the laparoscopic approach.

7 Ann Thorac Surg LOW AND UNGER 2005;80: OPEN REPAIR OF PARAESOPHAGEAL HERNIA 293 Table 10. Comparative Outcomes of Laparoscopic Series (Table 9) and Current Virginia Mason Medical Center Series of Open Paraesophageal Hernia Repairs Outcome However, visceral injuries are more commonly seen in laparoscopic operative series. When the combined results of the laparoscopic reports are compared with the current series (Table 10), the difference in length of stay is not as marked. In addition, the operative times are better and the open operation compares favorably with respect to visceral injuries, hernia recurrence, and mortality. In summary, paraesophageal hernia repair remains a challenging technical procedure. Elective repair is indicated in the majority of patients because urgent repair is associated with poor outcomes, but more importantly the impact of symptoms and other clinical issues on these patients has been underestimated. Results of open paraesophageal hernia repairs have continued to evolve in the same era that has seen the introduction of laparoscopic operations. Outcomes of open procedures are comparable to and, in some variables, better than those seen with the laparoscopic approach. Whether performed open or laparoscopically, these repairs should be preformed in experienced units to minimize morbidity and mortality. Hernia recurrence does occur in a portion of patients, and the clinical significance is yet to be conclusively determined, although recurrence is one of the most common reasons for the need for revisional operations. Quality-of-life factors routinely improve after open paraesophageal hernia repair. References Laparoscopic Reports ( ) Open Repairs Current Report ( ) Operation length (min) Hospital length of stay (days) Visceral injuries (%) Mortality (%) Hernia recurrence (%) Allen MS, Trastek VF, Deschamps C, Pairolero PC. Intrathoracic stomach: presentation and results of operation. J Thorac Cardiovascular Surg 1993;105: Hill LD. Incarcerated paraesophageal hernia: a surgical emergency. Am J Surg 1973;126: Skinner DB, Belsey RHR, Russell PS. Surgical management of esophageal reflux and hiatus hernia. J Thorac Cardiovascular Surg 1967;53: Pope CE. The quality of life following antireflux surgery. World J Surg 1992;16: Low DE, Ludwig DJ. Modified Hill repair for gastroesophageal reflux. In: Mastery of surgery, vol 1, 4th ed. Lippincott Williams &Wilkins, 2001: Low DE, Anderson RP, Ilves R, Riccciardelli E. Fifteen to twenty year results after Hill antireflux operation. J Thoracic Cardiovasc Surg 1988;98: Diaz S, Brunt LM, Klingensmith ME, Frisella PM, Soper NF. Laparoscopic paraesophageal hernia repair, a challenging operation: medium-term outcome of 116 patients. J Gastrointestinal Surg 2003;7: Pierre AF, Luketich JD, Fernando HC, et al. Results of laparoscopic repair of giant paraesophageal hernias: 200 consecutive patients. Ann Thorac Surg 2002;74: Jobe BA, Aye RW, Deveney CW, Domreis JS, Hill LD. Laparoscopic management of giant type III hiatal hernia and short esophagus: objective followup at three years. J Gastrointest Surg 2002;6: Gantert WA, Patti MG, Arcerito M, et al. Laparoscopic repair of paraesophageal hiatal hernias. J Am Coll Surg 1998;186: Wiechmann RJ, Ferguson MK, Naunheim KS, et al. Laparoscopic management of giant paraesophageal herniation. Ann Thorac Surg 2001;71: Mattar SG, Bowers SP, Galloway KD, Hunter JG, Smith CD. Long-term outcome of laparoscopic repair of paraesophageal hernia. Surg Endosc 2002;16: Perdikis G, Hinder RA, Filipi CJ, et al. Laparoscopic paraesophageal hernia repair. Arch Surg 1997;132: Horgan S, Eubanks TR, Jacobsen G, Omelanczuk P, Pellegrini CA. Repair of paraesophageal hernias. Am J Surg 1999;177: Swanstrom LL, Jobe BA, Kinzie LR, Horvath KD. Esophageal motility and outcomes following laparoscopic paraesophageal hernia repair and fundoplication. Am J Surg 1999;177: Williamson WA, Ellis FH, Streitz JM, Shahian DM. Paraesophageal hiatal hernia: is an antireflux procedure necessary? Ann Thorac Surg 1993;56: Geha AS, Massad MG, Snow NJ, Baue AE. A 32-year experience in 100 patients with giant paraesophageal hernia: the case for abdominal approach and selective antireflux repair. Surgery 2000;128: Altorki NK, Yankelevitz D, Skinner DB. Massive hiatal hernias: the anatomic basis of repair. J Thorac Cardiovascular Surg 1998;11: Myers GA, Harms BA, Starling JR. Management of paraesophageal hernia with a selective approach to antireflux surgery. Am J Surg 1995;170: Maziak DE, Todd TRJ, Pearson FG. Massive hiatus hernia: evaluation and surgical management. J Thorac Cardiovascular Surg 1998;115: Ellis FH, Crozier RE, Shea JA. Paraesophageal hiatus hernia. Arch Surg 1986;121: Martin TR, Ferguson MK, Naunheim KS. Management of giant paraesophageal hernia. Dis Esophagus 1997;10: Dahlberg PS, Deschamps C, Miller DL, Allen MS, Nichols FC, Pairolero PC. Laparoscopic repair of large paraesophageal hiatal hernia. Ann Thorac Surg 2001;72: Schauer PR, Ikramuddin S, McLaughlin RH, et al. Comparison of laparoscopic versus open repair of paraesophageal hernia. Am J Surg 1998;176: Hashemi M, Peters JH, DeMeester TR, et al. Laparoscopic repair of large type III hiatal hernia: objective followup reveals high recurrence rate. J Am Coll Surg 2000;190: Trus TL, Bax T, Richardson WS, et al. Complications of laparoscopic paraesophageal hernia repair. J Gastrointestinal Surg 1997;1: Wu JS, Dunnegan DL, Soper NJ. Clinical and radiologic assessment of laparoscopic paraesophageal hernia repair. Surg Endosc 1999;13: Mittal SK, Awad ZT, Tasset M, et al. The preoperative predictability of the short esophagus in patients with stricture or paraesophageal hernia. Surg Endosc 2000;14: Luketich JD, Raja S, Fernando HC, et al. Laparoscopic repair of giant paraesophageal hernia: 100 consecutive cases. Ann Surg 2000;232:

8 294 BOOK REVIEW Ann Thorac Surg OPEN REPAIR OF PARAESOPHAGEAL HERNIA 2005;80: Jobe BA, Horvath KD, Swanstrom LL. Postoperative function following laparoscopic Collis gastroplasty for shortened esophagus. Arch Surg 1998;133: Stylopoulos N, Gazelle GS, Rattner DW. Paraesophageal hernias: operation or observation? Ann Surg 2002;236: Treacy PJ, Jamieson GG. An approach to the management of oesophageal hiatus hernia. Aust NZ J Surg 1987;57: Low DE, Simchuk EJ. Effect of paraesophageal hernia repair on pulmonary function. Ann Thorac Surg 2002;74: Streets CG, DeMeester SR, DeMeester TR, et al. Excellent quality of life after Nissen fundoplication depends on successful elimination of reflux symptoms and not the invasiveness of the surgical approach. Ann Thorac Surg 2002;74: Thoracic Surgery Residents Association (TSRA) Resident Representatives President D. Michael McMullan, MD Vice-President Seenu Reddy, MD Immediate Past-President Grayson H. Wheatley III, MD TSRA Executive Council Resident Representatives The Society of Thoracic Surgeons Daniel J. Boffa, MD Cleveland Clinic Foundation John R. Mehall, MD University of Cincinnati American Association for Thoracic Surgery (AATS) Sanjeev Aggarwal, MD University of Michigan Sanjay Samy, MD Boston Medical Center Joint Council for Thoracic Surgery Education Anastasios Konstantakos, MD Beth Israel Deaconness Rakesh M. Suri, MD Mayo Clinic Organization of Resident Representatives of the American Association of Medical Colleges Parvez Sultan, MD Washington University, St. Louis Grayson H. Wheatley III, MD Arizona Heart Institute Thoracic Surgery Resident Review Committee (ACGME) Seenu Reddy, MD Emory University CTSNet Resident Section Editor Vinod Thourani, MD Emory University American Medical Association Resident/Fellows Section Eric Roselli, MD Cleveland Clinic Foundation AATS Web Page Committee Representative Ali Khoynezhad, MD Montefiore Medical Center 2005 by The Society of Thoracic Surgeons Ann Thorac Surg 2005;80: /05/$30.00 Published by Elsevier Inc