The American Surgeon™

In the era of endoscopic ultrasound-guided fine needle aspiration (EUS-FNA), a minimally invasive, safe, and accurate cytologic examination of a variety of intra-abdominal neoplasms has become possible. To assess the efficacy of EUS-FNA for diagnosis of primary pancreatic tumors, a 10-year retrospective review of a consecutive series of patients diagnosed between 1996 and 2005 was undertaken. Comparisons were made between early (1996–2000) and late (2001–2005) periods regarding diagnostic modalities used and the concordance of EUS-FNA cytology with macroscopic tissue histology. Although macroscopic biopsy diagnostic yield did not change over time, yield from EUS-FNA increased from 40 per cent to 95 per cent (P = 0.001). Because of improved accuracy of FNA cytology, only six per cent of tumors required additional macroscopic tissue histology in the late period versus 35 per cent in the early period (P = 0.001). There was 100 per cent concordance between the cytologic and histologic diagnoses in the late period versus only 33 per cent in the early period (P = 0.032). We conclude that 1) the frequency of pathologically diagnosed pancreatic tumors doubled over 10 years, 2) utilization of EUS-FNA significantly increased the accuracy of cytologic diagnosis, and 3) as a result, the need for macroscopic tissue biopsy for diagnosis of pancreatic neoplasms has been obviated.

Few studies have compared outcomes of right colectomy (RC) and left colectomy (LC) with respect to both benign and malignant disease. The objective of this study was to compare outcomes of RC versus LC for benign and malignant disease using a national administrative database of academic medical centers. Using International Classification of Diseases, 9th Revision diagnosis and procedure codes, data was obtained from the University HealthSystem Consortium Clinical Data Base for patients that underwent RC and LC for benign and malignant disease between 2002 and 2006. The main outcomes compared were demographics, length of hospital stay, observed to expected in-hospital mortality, complications, 30-day readmission, and mean cost. There were a total of 27,483 patients; 12,971 patients (47.2%) underwent RC. Compared with LC for benign disease, RC was associated with a shorter length of stay, lower overall complications, lower wound infections, lower 30-day readmissions, and lower cost. Compared with LC for malignant disease, RC was associated with lower overall complications, lower wound infections, and lower cost. In this analysis of academic centers, RC was associated with a lower length of stay, lower morbidity, and lower cost when compared with LC for benign and malignant disease.

Acute cholangitis is a life-threatening complication of biliary obstruction that is exacerbated by delays in diagnosis and treatment. Since the introduction of endoscopic retrograde cholangiography and endoscopic therapeutic modalities, few investigations have addressed admission prognostic indicators of adverse outcomes. A retrospective review of all patients with a diagnosis of acute cholangitis from 1995 to 2005 was performed. Primary endpoints were organ failure and death. One-hundred and seventeen patients met criteria for acute cholangitis. Only 49 (42%) had Charcot's triad and 3 (3%) had Reynolds’ pentad. One-hundred and four (89%) patients underwent biliary decompression, of which 79 (76%) were treated by endoscopic methods. There were 29 (25%) cases of organ failure and 9 (8%) deaths. The admission white blood cell (WBC) count (P = 0.0003) and total bilirubin (TBili) (P = 0.04) were statistically significant predictors of organ failure or death. With an admission of WBC ≥ 20,000 cells/mm³, the sensitivity, specificity, positive predictive value, and negative predictive value for organ failure and death were 50 per cent, 92 per cent, 63 per cent, and 88 per cent, respectively. A TBili of ≥10 mg/dL had sensitivity, specificity, positive predictive value, and negative predictive value of 56 per cent, 85 per cent, 21 per cent, and 96 per cent, respectively for predicting death. Admission WBC ≥ 20,000 cells/mm³ and TBili ≥ 10 mg/dL are selective predictors of adverse outcomes in acute cholangitis.

No standardized approach exists for laparoscopic Roux-en-Y gastric bypass (LRYGB). At a newly instituted bariatric surgery program, four experienced laparoscopic surgeons used the systematic and evidence-based approach consisting of multidisciplinary preoperative evaluation, screening, and education; standardized operative technique; inpatient clinical pathway; and close postoperative follow-up. The outcomes were subsequently analyzed to determine if this approach improved the morbidity and mortality. From January 2003 to June 2006, 835 consecutive LRYGBs were performed. The patient population was 85 per cent women with a mean body mass index (BMI) of 50.4 kg/m² (range 33–96 kg/m²). The mean age was 44 (range 15–67). Sixty-two per cent of the patients had previous abdominal or pelvic operations. The conversion rate to open surgery was 0.2 per cent. The average length of hospital stay was 2.6 days (range 2–13 days). There were no anastomotic leaks or deaths. The 30-day readmission and re-operation rates were 3.2 per cent and 1.8 per cent, respectively. The incidence of anastomotic stricture, marginal ulcer, bleeding, pulmonary embolism, and internal hernia was 0.8 per cent, 3.5 per cent, 4.2 per cent, 0.1 per cent, and 0.4 per cent, respectively. A systematic and evidence-based approach to the LRYGB by experienced laparoscopic surgeons resulted in a lower incidence of complications when compared with the published results from other comparable institutions.

Controversy remains whether patients with body mass index (BMI) ≥ 50 kg/m² experience more complications after laparoscopic Roux-en-y gastric bypass (LRYGB) than those with a lower BMI. Whether BMI ≥ 50 kg/m² is associated with specific complications remains unknown. Charts of 152 patients who underwent LRYGB were reviewed. Logistic regression was used to determine whether high BMI is associated with minor or major complications. Overall, there was a trend that major complications occurred more frequently in patients with BMI ≥ 50 compared with BMI < 50, (30.4% vs 19.8%, P = 0.138). Major bleeding complications occurred in 16.1 per cent of high BMI patients as compared with 5.2 per cent with lower BMI (P = 0.025). Multivariate regression found that BMI ≥ 50 was associated with higher odds of a major technical complication (OR = 2.73, P = 0.04), particularly for bleeding complications (odds ratio [OR] = 5.59, P = 0.01). Male gender was also associated with higher odds of a major technical complication (OR = 3.43, P = 0.04). These results suggest that high BMI patients may be better candidates for other types of weight loss surgery, such as staged procedures, and that surgeons early in their career should operate on patients with lower BMI.

This study was undertaken to identify characteristics of residents who left their training program before the end of the program. A survey was sent to 248 Program Directors in the United States, after institutional review. Anonymous responses were received from 27 (11%) programs. Data was received on 166 residents, including 111 males and 55 females. The group included 146 categorical residents and 20 preliminary residents. Of these, 60 residents left in the middle of the year and 105 left at the end of the year. Of the 164 residents for whom data was reported, 30 (21%) used counseling services. Of 110 residents who left before 2004, 54 (49%) left to choose another specialty. In 2004, of 25 residents who left, 13 (52%, P > 0.05) pursued training in another specialty, and in 2005 of 31 residents who left, 23 (74%, P < 0.025) chose another specialty. Significantly more residents who started a career in surgery after 2004 left to train in another specialty. This may be due to implementation of work hour restrictions in 2003 leading residents to enter surgery who would not have done so previously. Strategies for better retention of matched residents are necessary to reverse this worrisome trend.

Malignant phyllodes tumor (MPT) is a rare breast malignancy. Because of the scarcity of the disease, there are no evidence-based treatment or follow-up guidelines established. This study evaluated the survival of MPT patients to create recommendations for management. We identified 752 cases of malignant phyllodes tumors in the California Cancer Registry from the years 1988 to 2003. Relative survival was determined using Berkson-Gage life table analysis which was then compared with the relative survival of nonphyllodes breast cancer patients. For MPT patients, the relative annual survival at 1 year was 94 per cent and at 10 years was 99.6 per cent. Thus, after 10 years, these patients are no more likely to die than the general population. At 10 years, the relative cumulative survival of the MPT patients was 87.4 per cent, whereas the nonphyllodes breast cancer patients had only a 57.2 per cent relative cumulative survival. MPT patients with localized disease had a higher 10-year relative cumulative survival than those with regional disease (90.9% vs 61.5%, P < 0.001). MPT has a good prognosis, particularly in patients with localized disease. After 10 years, MPT patients have no increased mortality relative to the general population. Clinicians should plan these patients’ follow-up accordingly.

The objective of our study is to assess the utility of breast Magnetic Resonance Imaging (MRI) when used for indications other than those published in peer-reviewed studies. A retrospective chart review was conducted of the records of 588 women who underwent both mammography and breast MRI. Patients excluded from the study were those who had breast MRI for accepted indications based on published peer-review studies. Included on the study were the remaining 122 patients. An evaluation was then made in each case as to whether the MRI finding caused a change in the patient's management. In this review, subject age ranged from 27- to 85-years-old. The mean age of the sample was 54.5 years. Of the positive MRI results, 29 (27.7%) had additional findings. There were 25 (20.3%) subjects with a treatment change and 97 (79.5%) without. In conclusion, breast MRI affected the clinical management in 25 (20.3%) of 122 patients. The majority of the 25 patients have invasive ductal carcinoma, followed by ductal carcinoma in situ. We believe this is a significant percentage positively affected by the additional use of breast MRI. We suggest that indications for the use of breast MRI in addition to traditional breast imaging should include all patients with invasive ductal carcinoma.

Positive margins and the resulting multiple operations are a major problem for breast conservation therapy. Accurate assessment of intraoperative tumor margins can limit multiple re-excision procedures. Intraoperative touch preparations have been used in the past but can be difficult to interpret without an experienced cytopathologist. The objective of this study is to examine the reliability of enhanced intraoperative touch preps (EIOTP) compared with final pathologic margins. We prospectively performed EIOTP on 20 tumors in women undergoing breast conservation therapy. Six margins and the main tumor were touched onto poly-L-lysine coated slides. The slides were stained with anti MUC1 and anti-E-cadherin antibodies, and Hoechst nuclear stain. A parallel set of slides were stained with hematoxylin and eosin for comparison. The EIOTP results were compared with pathologic interpretation of paraffin embedded permanent sections. A total of 120 margins underwent EIOTP in 20 patients. We found a sensitivity equal to 80 per cent, specificity 100 per cent, positive predictive value 100 per cent, and negative predictive value 99 per cent. EIOTP in conjunction with MUC-1 and E-cadherin by immunofluorescence is a sensitive and highly specific mechanism to identify cancer cells at breast tissue margins. The immunofluorescence stains may help the pathologist to identify cancer cells in fresh breast tissue and limit breast re-excisions in the future.

With the increasing usage of neoadjuvant chemotherapy (NAC) in locally advanced breast cancer (LABC), there is the need to investigate the routine axillary node dissections performed in this group of patients. Controversy exists about the utility of sentinel node biopsy (SNB), either before or after NAC. With the addition of trastuzumab in the treatment of Her2/neu-positive LABC patients, the validity of SNB in this subset population needs to be investigated. A retrospective study of 20 patients who underwent NAC for LABC was undertaken. The pathology of the axillary nodes, sentinel nodes, and primary tumor after neoadjuvant chemotherapy were examined. Twenty patients underwent NAC with doxorubicin and cyclophosphamide, followed sequentially by paclitaxel and carboplatin, with or without trastuzumab based on Her2/neu status. Post chemotherapy, 20 patients underwent mastectomy or lumpectomy with SNB with axillary node dissections. The overall accuracy of SNB was 95 per cent with a false-negative rate of 14 per cent (1/7). In Her2/neu-positive patients, overall accuracy was 100 per cent (8/8) and a false-negative rate of zero per cent. Sentinel node biopsy is a viable option in patients who have undergone NAC. Her2/neu-positive patients who had undergone NAC with trastuzumab had comparable accuracy for sentinel node biopsy in predicting axillary node status.

Sentinel lymph node biopsy (SLNB) is now an established method of axillary staging in patients with breast cancer. However, the augmented breast poses an interesting challenge to this procedure. We hypothesized that SLNB is feasible in patients with augmented breasts who subsequently develop breast cancer. A retrospective study was performed from 1995 to 2006. Ten patients with augmented breasts underwent breast conservation therapy with SLNB. Sentinel lymph nodes were identified in all 10 patients. Three patients had positive sentinel nodes. Two patients proceeded to axillary lymph node dissection (ALND), and one declined. The subsequent ALND were negative for metastatic cancer. Seven patients had negative sentinel nodes. One patient with a negative sentinel node underwent ALND with all nodes negative for metastasis. Two patients were lost to follow-up. Of the remaining eight patients, the mean duration of follow-up was 71 months. None of these patients had evidence of axillary recurrence or distant metastasis at time of last follow-up. SLNB is a feasible method of axillary node staging in patients who have undergone augmentation mammoplasty who subsequently develop breast cancer. Further studies are needed to better determine the accuracy of lymphatic mapping in this patient population.

Columnar cell lesion with atypia (CCLA) is a newly recognized pathologic entity seen in breast specimens. The breast cancer risk associated with this finding is unclear, although CCLA had been found adjacent to both in situ and invasive carcinomas, but the incidence is unknown. Breast specimens from patients with a columnar cell lesion were reviewed by a pathologist for atypia. Twenty-one specimens with CCLA were identified [core biopsy (8), excisional biopsy (11), and simple mastectomy (2)]. Six of eight specimens with CCLA on core had adjacent abnormal pathology: infiltrating ductal carcinoma (IDC)/lobular carcinoma in situ (LCIS) (1), ductal carcinoma in situ (DCIS)/LCIS (1), DCIS (1), LCIS (1), and papillomatosis (2). Five of 11 specimens with CCLA on excisional biopsy had adjacent abnormal pathology: IDC (3), DCIS/LCIS (1), and atypical ductal hyperplasia/papilloma (1). Two of two simple mastectomy specimens had CCLA associated with IDC (1) and DCIS (1). Overall, abnormal pathology was found adjacent to CCLA in 62 per cent of specimens (13/21). Breast pathologic specimens containing a columnar cell lesion should be carefully examined for atypia. Surgical excision is warranted for CCLA found on core biopsy. The future risk of breast cancer based on the finding of CCLA alone requires further investigation.

The technical aspects of laparoscopic hepatic resection have evolved rapidly. The key to any approach is establishing a reliable method to prevent or control hemorrhage during parenchymal transection. Although combining a hand-assist technique with laparoscopy allows improved control of bleeding risk, this requires the addition of a hand-port incision. The development of novel devices that can be used to safely divide liver parenchyma laparoscopically may lessen the need for hand-assist. Here, we report a series of laparoscopic hepatic resections that were attempted without the use of hand-assistance (completely laparoscopic). Resections were performed using saline-cooled cautery (Tissue-Link Endohook) and/or hydrodissection (Erbe Helix Hydrojet). Fifteen laparoscopic hepatic resections were attempted by a single surgeon from 2002 to 2006. In each case, a nonanatomic, completely laparoscopic approach was attempted. Patients with lesions at the hepatic dome or those requiring lobectomy or hilar dissection were excluded. Fourteen of 15 cases (93%) were accomplished completely laparoscopically, while one patient required placement of a hand port. Resected tumors averaged 3.9 cm diameter. There were no bile leaks and no patient required transfusion. Average length of stay was 4.1 days (range 1–5). Complications included ileus (1) and atrial fibrillation (1). In six patients with malignancies, margins were negative and there have been no local or port recurrences. This report demonstrates the feasibility of completely laparoscopic hepatic resection using novel devices for parenchymal transaction. Hand-assist techniques remain useful as a salvage strategy or for larger resections.

Despite advances in perioperative care and operative techniques, urgent colorectal operations are associated with higher morbidity and mortality. To evaluate our rate of complications in elective and urgent colorectal operations, we performed retrospective chart review of 209 consecutive patients who underwent colorectal resection between 1998 and 2002 at Harbor-UCLA Medical Center. One hundred, forty-three (71%) patients underwent elective colorectal resection. A total of 19 (13.3%) complications occurred in the elective group, compared with 24 (38.1%) in the urgent group (P = 0.003). Both right-sided and left-sided operations were associated with higher incidence of complications when performed urgently. Wound infection occurred in 7.7 per cent of patients undergoing an elective operation and 14.3 per cent in an urgent setting (P = 0.21). Intra-abdominal abscess occurred in 1.4 per cent of patients undergoing elective operation, compared with 11.1 per cent in the urgent operation group. Four (1.9%) patients developed wound dehiscence, 1 in elective and 3 in the urgent group (P = 0.09). Anastomotic leak occurred in 1.9 per cent of patients, 2 in each group (P = 0.6). There were six deaths, 3 in elective and 3 in urgent cases (P = 0.4). Urgent operation of the colon and rectum is associated with higher incidence of complications. Both right- and left-sided resections have a higher complication rate when performed in a nonelective setting.

The best way to evaluate the colon for both diagnosis of symptoms and surveillance is colonos-copy. However, access to colonoscopy is often restricted. Our objective was to assess the anatomic distribution and stage at presentation of colorectal cancer (CRC) in a county hospital population, the prevalence and distribution of CRC in younger patients, and the utility of flexible sigmoid-oscopy for early diagnosis of left-sided cancers in this population. We performed a retrospective chart review of 151 patients who underwent colorectal resection from 2001 to 2003. Overall, 66.9 per cent of patients underwent resection for left-sided CRC. Forty-two (27.8%) of 151 were under age 50. In patients over 50, 66.1 per cent were found to have left-sided CRC compared with 69 per cent of patients under 50. Fifty per cent (50.3%) of patients had stage III or IV (advanced) disease. Forty-nine and a half per cent of patients over 50 and 52.3 per cent under 50 had advanced disease. Forty-eight and a half per cent of patients with left-sided CRC had advanced stage disease compared with 54% of patients with right-sided CRC. In patients under 50, the rates were 55.2 per cent and 46.1 per cent respectively. Two-thirds of the CRC occurred in the left side of the colon in both older and younger population. Flexible sigmoidoscopy should be considered as an early tool in the diagnosis of CRC.

Unplanned readmission (UR) is considered to be an index of quality surgical care. We examined whether any perioperative factor was associated with UR after colectomy for ulcerative colitis (UC) or indeterminate colitis (IC). Patients undergoing a two-stage or three-stage ileal pouch-anal anastomosis were included. Patient, disease, and surgical factors were collected. UR occurring within 30 days of hospital discharge was assessed. The 202 study patients had a median age of 38 years. Median body mass index was 22. There were 130 (64%) UC patients and 72 (36%) IC patients. Indications for surgery were medically refractory disease (n = 176, 87%) and dysplasia/cancer (n = 26, 13%). Preoperative medical therapy included steroids alone in 25 patients and steroids combined with other immunomodulators in 151 patients. A two-stage and three-stage ileal pouch-anal anastomosis was used in 146 (72%) and 56 (28%) patients, respectively. Median white blood cell count before discharge was 8600 cells/mm³. Median length of stay after surgery was 7 days. Complications before discharge were observed in 28 patients (14%). Thirty-eight patients (19%) had a UR. No preoperative or surgical factor was associated with UR. Although UR occurs frequently (19%) after colectomy for UC or IC, it cannot be predicted.

Several studies report sensitivity and specificity of abdominal computed tomography scans (CT) for the evaluation of acute appendicitis as high as 98 per cent. Despite increased utilization of CT, the rate of negative appendectomy has remained constant at 10 to 20 per cent. The objective of this study was to assess the effectiveness of CT in the evaluation of acute and perforated appendicitis in an academic community-based setting. A retrospective review of 550 patient charts with International Classification of Diseases-9 (ICD-9) codes for acute and perforated appendicitis from January 2002 to October 2005 was performed. Sensitivity of CT was 87 per cent with a positive predictive value of 92 per cent. Specificity was 42 per cent with a negative predictive value of 29 per cent. Negative appendectomy rates were similar with or without CT (11% vs 13%, respectively). Our data suggests that CT used liberally in everyday practice in a community-based setting to evaluate acute appendicitis may not have as strong of a diagnostic value as those used in protocol-driven research studies. Further prospective studies are needed to formulate criteria to better delineate the role of CT in the evaluation of acute appendicitis.

In the six centuries since mesenteric ischemia was first described, multiple factors have been investigated as predictors of bowel viability with little consensus. We retrospectively examined all cases of exploratory laparotomy for suspected bowel ischemia over an 8.5-year period. Patients were grouped into those who had a “positive laparotomy” with findings of bowel ischemia, bowel gangrene, or both (PL) and those who had a “negative laparotomy” with no evidence of compromised bowel (NL). Of the 114 patients, 86 (75%) were in the PL group and 28 (25%) in the NL group. The significant differences between the two groups were the higher prevalence of females in the PL group versus the NL group [71% vs 50% (P = 0.04)] and the younger age in the NL group versus the PL group [64 ± 19 vs 76 ± 14 (P = 0.0002)]. The groups did not significantly differ with respect to preoperative comorbidities and results of laboratory studies. Two patients in NL group had pneumatosis intestinalis on abdominal CT. One patient in PL group had a negative visceral angiogram. There was no difference in mortality between the groups. No single preoperative study reliably predicted positive findings at laparotomy in our series.

Patients who are obese are believed to be at greater risk of developing intraoperative and postoperative complications than their nonobese counterparts. Many studies have shown that there is a higher prevalence of obesity among the Veterans Affairs patient population. We performed a retrospective review of 941 patients presenting to a single Veterans Affairs Medical Center. We aimed to determine the incidence of obesity among the Veterans Affairs Medical Center general surgery patient population as well to compare the frequency of postoperative complications between patients who are obese and nonobese patients undergoing elective general surgery. Body mass index was calculated for all patients; of the 941 patients seen in the clinic, 547 underwent elective surgery. Thirty-three per cent of all clinic patients had a body mass index greater than 30 kg/m². Twenty-eight per cent of patients who underwent surgery had a body mass index greater than 30 kg/m². Postoperative complications developed among 5.5 per cent of all surgical patients; 23.3 per cent were obese and 76.7 per cent were nonobese. There was no statistically significant difference between these two groups (P = 0.54). This study illustrates the increased prevalence of obesity among the Veterans Affairs Medical Center general surgery patient population. In addition, our study suggests that obesity is not an independent risk factor for postoperative complications in patients undergoing elective general surgery.

Of the approximately one in 1000 pregnant women who develop the syndrome of hemolysis, elevated liver enzymes, and low platelets (HELLP), 2 to 3 per cent develop hepatic complications, including liver failure for which liver transplantation (LT) may be required. Between February 1, 1984, and December 31, 2006, eight women without a history of liver disease underwent LT for complications of HELLP syndrome. All received cadaveric grafts with a mean interval from delivery to LT of 7 days. The mean admission Child-Turcotte-Pugh score was 13.1 (class C), and the mean model for end-stage liver disease score was 40. Manifestations of liver failure included encephalopathy (seven patients), renal failure (four), disseminated intravascular coagulation (three), and respiratory failure (one). There were no intraoperative deaths. Complications of LT included biliary leaks (three patients), reoperation (three), and retransplantation (two). There was one death from sepsis on postoperative day 91 and one death from cholangitis/sepsis more than 5 years postoperatively. After LT, 1-, 5-, and 10-year patient survival rates were 88 per cent, 88 per cent, and 65 per cent; 1-, 5-, and 10-year graft survival rates were 64 per cent, 64 per cent, and 48 per cent. This is the largest single-center report of LT for HELLP. Early recognition and transfer to a transplant center will yield best results with this challenging complication of pregnancy.

Alterations in regional tissue perfusion may precede global indications of shock. This study compared regional tissue oxygenation saturation (S_tO₂) using near-infrared spectroscopy with standard hemodynamic and biochemical variables in 40 patients undergoing cardiopulmonary bypass (CPB). Mean arterial pressure, cardiac output, oxygen delivery, arterial blood gases, and lactate were recorded at specific intervals during surgery. Data were organized by stage of procedure, and the relationship of S_tO₂ to established parameters was investigated. With initiation of CPB, S_tO₂ declined by 12.9 per cent (standard deviation ± 14.75%) with a delayed increase in lactate from 0.9 (interquartile range [IQR], 0.6–1.5) mmol/L to 2.3 (IQR, 1.8–2.5) mmol/L. The minimum S_tO₂ value preceded the maximum lactate level by an average time of 93.9 (standard deviation ± 86.3) minutes. Additionally, a decrease in S_tO₂ corresponded with an increase in base deficit of 4.84 (standard deviation ± 2.37) mEq/L over the same period. Calculated oxygen delivery decreased from a baseline value of 754 (IQR, 560–950) mL/min to 472 (IQR, 396–600) mL/min with initiation and maintenance of CPB. For patients undergoing CPB, S_tO₂ is a reliable, noninvasive monitor of perfusion, which correlates well with oxygen delivery and identifies perfusion deficits earlier than lactate or base deficit.

There are no independent computed tomography (CT) findings predictive of elevated intracranial pressure (ICP). The purpose of this study was to evaluate brain density measurement on CT as a predictor of elevated ICP or decreased cerebral perfusion pressure (CPP). A prospectively collected database of patients with acute traumatic brain injury was used to identify patients who had a brain CT followed within 2 hours by ICP measurement. Blinded reviewers measured mean density in Hounsfield Units (HU) within a 100-mm² elliptical region at four standardized positions. Brain density measurement was compared for patients with an ICP of 20 or greater versus less than 20 mm Hg and CPP of 70 or greater versus less than 70 mm Hg. During a 2-year period, 47 patients had ICP monitoring after brain CT. Average age was 40 ± 18 years old; 93.6 per cent were male; mean Injury Severity Score was 25 ± 10; and Glasgow Coma Scale was 6 ± 4. There was no difference in brain density measurement for observer 1, ICP less than 20 (26.3 HU) versus ICP 20 or greater (27.4 HU, P = 0.545) or for CPP less than 70 (27.1 HU) versus CPP 70 or greater (26.2, P = 0.624). Similarly, there was no difference for observer 2, ICP less than 20 (26.8 HU) versus ICP 20 or greater (27.4, P = 0.753) and CPP less than 70 (27.6 HU) versus CPP 70 or greater (26.2, P = 0.436). CT-measured brain density does not correlate with elevated ICP or depressed CPP and cannot predict patients with traumatic brain injury who would benefit from invasive ICP monitoring.

In 1994, Bickell et al. published a prospective study recommending restricting prehospital intravenous fluids (IVF) to less than 100 cc in patients with penetrating truncal injuries and field hypotension, reporting a 30 per cent mortality with IVF restriction and a 38 per cent mortality with liberal IVF use. However, since this study, few papers have investigated whether emergency medical systems (EMS) adhere to these IVF guidelines. The purpose of this study was to determine whether a policy of IVF restriction is being followed and whether the volume of prehospital and emergency department (ED) IVF affects outcome in patients with penetrating truncal injury and field hypotension at a Level I trauma center in Los Angeles County. A retrospective analysis of a trauma database from 1998 to 2005 of all patients with penetrating truncal injury and field hypotension (systolic blood pressure less than 90 mm Hg) was performed. Multiple variables, including originating EMS agency, mechanism of injury, transport time, Injury Severity Score, field and ED vital signs, and IVF volume infused, complications, and mortality were compared. One hundred ninety-four patients with a median age of 26 years with penetrating truncal injury and field hypotension were analyzed. The most common mechanisms of injury were gunshot (73%) and stab (22%) wounds. The median field systolic blood pressure was 80 mm Hg. The median transport time was 11 minutes. The median prehospital IVF was 500 cc with only 25 per cent receiving less than 100 cc of IVF. There were no differences in the amount of IVF administered by the degree of field hypotension or by originating EMS agency. Median ED IVF was 1000 cc. The overall mortality rate was 25 per cent. When a comparison was made of those receiving less than 100 cc prehospital IVF in comparison to those receiving greater than 100 cc, there were no differences detected with respect to median age, systolic blood pressure, Injury Severity Score, transport time, or morbidity rate. The mortality rate was 21 per cent in the group that received greater than 100 cc of IVF in comparison to a 37 per cent mortality rate in the group that received less than 100 cc IVF (P = 0.04). On multivariate analysis, after adjusting for Trauma Injury Severity Score, there were no differences in survival by the amount of prehospital or ED IVF administered. It appears that the recommendations of IVF restriction for patients with penetrating truncal injuries and field hypotension are not being followed by Los Angeles County EMS. There were no differences in survival with respect to the amount of prehospital or ED IVF. Given the retrospective nature of this study, further investigation is needed to define the role of prehospital IVF resuscitation in these patients.

Trauma patients are thought to be at high risk for iatrogenic retained foreign bodies (RFBs). The objective of this study was to evaluate this incidence. All cases of RFB after cavitary trauma surgery were identified by review of Morbidity and Mortality reports at a Level 1 trauma center from January 1998 to December 2005 and confirmed by the Octagon Risk Management System. Over 8 years, 10,053 trauma operations were performed (2075 laparotomies, 377 thoracotomies, and 74 sternotomies). Three cases (0.1%) of RFB (all sponges) occurred during one single-stage and two damage control laparotomies. The counts were correct before definitive closure in two of three cases. No postoperative x-rays were obtained in any of the cases. RFB diagnosis occurred between days 3 and 9, one on a routine chest x-ray and the other two on abdominal computed tomography scans during a septic workup. Four-month to 8-year follow up documented one pleural effusion and one abscess resulting from the RFB. Iatrogenic RFBs after emergent cavitary trauma surgery occur at a rate of 0.12 per cent and are associated with significant morbidity. In addition to standard preventive strategies, in emergent cases with risk factors such as requiring damage control, before final cavity closure, even with a correct sponge count, radiographic evaluation is warranted.

The evaluation of the abdomen in patients with spinal cord injury (SCI) is challenging for obvious reasons. There are very little data on the incidence and complications of patients who sustain SCI with concomitant intraabdominal injury (IAI). To determine the incidence and outcomes of IAI in blunt trauma patients with SCI, a trauma registry and record review was performed between January 1998 and December 2005. Baseline demographic data, Injury Severity Score, and associated IAI were collected. Two groups were established and outcomes were analyzed based on the presence or absence of IAI. Intraabdominal and hollow viscus injures were found in 15 per cent and 6 per cent, respectively, of 292 patients with blunt SCI. The presence of intraabdominal injury varied according to the level of the SCI: 10 per cent of cervical, 23 per cent of thoracic, and 18 per cent of lumbar SCI. The overall mortality was 16 per cent. The presence of intraabdominal injury was associated with longer intensive care unit length of stay (13 versus 6 days, P < 0.01), hospital length of stay (23 versus 18 days, P < 0.05), higher complication rate (46% versus 33%, P = 0.09), and higher mortality (44% versus 11%, P < 0.01) when compared with patients with SCI without IAI. Intraabdominal injuries are common in blunt SCI. Liberal evaluation with computed tomography is necessary to identify injuries early.

Management of lower extremity venous trauma using repair or ligation has been an area of controversy during the past decades. However, in unstable patients or if primary repair is technically impossible as a result of extensive disruption of the vein, ligation is recommended. This study investigated the effects of venous ligation on major veins in the lower extremities when primary repair is impossible as a result of extensive laceration of the vein. Between January 2001 and April 2004, 63 patients with Grade III and IV venous injuries were observed prospectively. Compression ultrasonography was performed postoperatively on the fifth day, once before discharge, and at the 3-month visit to assess deep vein thrombosis (DVT) and the patency of arterial repair. If DVT was present, the patient was given an oral anticoagulant (warfarin Na) for 3 months (international normalized ratio, 2.0–3.0), and Class II compression stockings (Sigvaris-212, Ganzoni, Switzerland) were used for 1 year. Follow-up visits occurred at 1, 3, 6, and 12 months and at 6-month intervals thereafter. Combined arterial and venous injuries were present in 50 (79.4%) patients and pure venous injuries were present in 13 (20.6%) patients. DVT developed in 49 patients (77.7%; postoperative n = 37 [58.7%], late n = 12 [19%]). Three arterial restenoses (4.7%) and one pseudoaneurysm (1.6%) of the superficial femoral artery developed. Five early (prophylactic) and two late (compartment syndrome) fasciotomies were performed. Postoperative edema was seen in 56 (88.8%) patients and wound infection was seen in 19 patients (30.1%; n = 18 superficial, n = 1 deep). Two amputations (3.2%) were performed. One patient (1.7%) died as a result of irreversible shock. After a median of 18 months, 25 patients were classified with Clinical Etiology, Anatomy, Pathology classification: 10 legs C-0, seven legs C-2, and eight legs C-3. No severe postthrombophlebitic syndrome was observed. Early leg swelling after venous ligation was the most common morbidity. We observed no significant sequelae of chronic venous insufficiency, and venous ligation had no detrimental effect on associated arterial repair. In cases of DVT, anticoagulation with low-molecular-weight heparin and oral anticoagulants should begin immediately and continue for 3 months along with compression stocking support for 1 year.

A retrospective review was conducted to analyze the effect of methamphetamine use in trauma patients. Charts of all trauma patients admitted to Kern Medical Center from January 1, 2003, to January 5, 2006 (36 months) were analyzed for length of stay, intensive care unit (ICU) admission rate and number of ICU days, ventilator days, and mortality. Results were compared in patients testing positive for methamphetamine (M+) with those who tested negative (M-). Data were then stratified according to six Injury Severity Score (ISS) groups: 1–5, 6–10, 11–15, 16–20, 21–25, and 26–30. A total of 4759 patients were admitted to the trauma unit, 971 of whom had available urine toxicology results. Six hundred seventy-four tested M- and 297 tested M+. There were no differences in total ICU days or ventilator days in the M+ versus M- patients. There was a higher incidence of assault in the M+ group (P = 0.0001). A trend toward decreased mortality was noted in M+ patients (P = 0.0778). ISS subset analysis demonstrated an increased ICU admission rate in M+ patients in ISS group 1–5 (P = 0.0002). There was also an increased length of stay in M+ patients within the ISS 6–10 group (8 versus 5 days, respectively, P = 0.015).

Standard formulas for predicting postoperative forced expiratory volume in 1 second (po-FEV₁) do not consider bronchi obstructed by tumor or chronic obstructive pulmonary disease, e.g., Formula 1 [ppo-FEV₁ = (pre-opFEV₁) x (# segments remaining)/(# of total segments)] whereas Formula 2 [ppo-FEV₁ = (pre-opFEV₁) x (# segments remaining)/(# of total unobstructed segments)] does. A retrospective chart review was conducted to determine accuracy of predicting po-FEV₁, at a comprehensive cancer center. Predicted po-FEV₁ was calculated using different formulas and analyzed using regression analysis and Pearson correlation. We found good correlation between po-FEV₁ and predicted po-FEV₁ using Formulas 1 and 2. In patients with tumor airway obstruction or chronic obstructive pulmonary disease, predictive accuracy decreased for both formulas. Prediction of FEV₁ in patients undergoing pulmonary resection was generally accurate, but major errors were observed in some cases; therefore, better predictive formulas are needed in patients with airway obstruction by tumor or chronic obstructive pulmonary disease.

An elevated thyroglobulin (Tg) level after total thyroidectomy for differentiated thyroid cancer is often associated with disease recurrence. ¹³¹I-whole body scans (¹³¹I-WBS) and cross-sectional imaging are commonly used to localize occult metastases in these patients. Localizing disease when ¹³¹I-WBS are negative and cross-section imaging is equivocal remains a challenge. The medical records of 12 patients with thyroid cancer undergoing positive positron emission tomography (PET) scans for ¹³¹I-WBS-negativeTg elevations or the presence of anti-Tg antibodies were identified and charts were reviewed in a retrospective fashion. All had been treated with total thyroidectomy and ¹³¹I ablation in the past. Computed tomography, magnetic resonance imaging, or ultrasound studies revealed suspicious lesions in eight patients. All 12 patients underwent resection of the PET-positive lesions. All resections were positive for thyroid cancer in the regions predicted by the positive PET scan. All nine (100%) patients with elevated preoperative Tg levels experienced a reduction in Tg level after resection. PET scans accurately predict the presence of recurrent thyroid cancer when ¹³¹I-WBS are negative. PET scans should be considered in the follow up of ¹³¹I-WBS-negative patients with thyroid cancer who are suspected of having recurrent disease.

We conducted a retrospective tumor registry review of a 36-year experience in a university center and identified 10 patients with duodenal lymphoma (five localized, five disseminated). Histologic types included diffuse large B-cell in four patients, mucosa-associated lymphoid tumor in three, and Hodgkin, follicular, and unclassified (one each). Treatments included chemotherapy in four patients, radiation therapy (RT) in two patients, Helicobacter pylori treatment in two, and observation in one. Five patients underwent operations (emergent in two, elective in three) for indications including massive bleeding in two patients, obstruction in two, or both in one. Survival for surgical group was 25 per cent at 1 year. One-year survival for nonsurgical group was 100 per cent, and all nonoperated patients lived at least 5 years, except for one who is alive 2 years after diagnosis. Surgical patients were younger and had more advanced lesions and less favorable cell types. When operation is required for bleeding or obstruction from secondary tumors in younger patients with disseminated disease, surgical challenges are formidable and survival is very limited. Tumors of less aggressive histology have far better prognosis.

Direct invasion of colorectal adenocarcinoma into adjacent structures occurs frequently, but only rarely is the duodenum involved. This study was undertaken to assess the safety and efficacy of en bloc resection of locally advanced right colon carcinoma invading the duodenum. A retrospective review of 49 patients with locally advanced colon cancer, surgically managed between 2000 and 2005, was performed. Forty-six patients underwent en bloc resection of colon and adjacent organs not involving the duodenum. Three patients with duodenal invasion underwent en bloc partial duodenectomy. The mean operative blood loss, length of stay, postoperative morbidity, and mortality compare favorably between these two groups of patients. Of the 46 patients with en bloc resection of other organs, 27 are alive at 12 to 60 months follow up. Two patients with duodenal invasion are alive without recurrence at 15 and 20 months follow up. En bloc resection of colon cancer invading the duodenum can be performed safely because morbidity and mortality rates are comparable to those attending extended resections of other locally advanced colon carcinomas. Overall survival in patients who underwent surgery with curative intent justifies en bloc duodenal resection in selected patients.

Light-assisted stab phlebectomy (LASP) is a technique that we have developed for treatment of lower extremity varicose veins. It combines powered and stab phlebectomy with minimally invasive surgical instruments to optimize intraoperative visualization of varicose veins and to remove branch veins with minimal bruising, hematoma, or pain. The early experience with this technique has not been reported. We conducted a retrospective review of all patients who underwent LASP to characterize the associated perioperative experience and to identify postoperative factors that indicated a successful outcome. From 2004 to 2006, LASP was performed on 268 limbs to remove lower extremity branch veins. The majority of patients were women (75%), and the mean age was 52 years. A group of 184 patients underwent LASP in combination with another procedure, and 49 patients underwent LASP alone. Nearly all patients had conscious sedation for anesthesia (99%); operating room time averaged 44 minutes, and blood loss averaged 88 cc. Immediate postoperative complications were infrequent, occurring in 10 per cent of patients, and included unresected or missed veins, hematoma, and cellulitis. LASP provides improved visualization of branch veins and allows varicose veins to be removed with a short operating room time and minor postoperative complications.

Little is known about ethnic differences among patients with gallstone pancreatitis in the United States. The purpose of this study was to compare Hispanic and non-Hispanic patients with gallstone pancreatitis with regard to severity of disease, level of care required, length of hospital stay, and clinical outcomes. A retrospective cohort study of 198 consecutive patients with gallstone pancreatitis was performed from 2003 to 2005. Overall, 161 patients were Hispanic and 37 were non-Hispanic. The average age of Hispanic patients was 41 years versus 47.5 years in the non-Hispanic group (P = 0.02). Only 16 (10%) Hispanic patients had a Ranson score of 2 or greater versus nine (24%) of the non-Hispanic group (P = 0.03). Only 39 (24%) Hispanic patients were admitted to an intensive care unit or stepdown unit versus 17 (46%) of the non-Hispanic group (P = 0.01). Hispanic patients underwent cholecystectomy at an average of 5.8 days after admission versus 6.6 days for non-Hispanic patients (P = 0.07). There was a 4 per cent complication rate and a 2 per cent readmission rate overall with no statistically significant differences between the two groups and no mortality. The majority of Hispanic patients with gallstone pancreatitis have a benign disease process, presenting at a younger age, with less severe disease that infrequently requires intensive care unit admission.

There is an ongoing debate regarding the optimal operation for chronic pilonidal disease. The current study analyzes the outcome of wide local excision (WLE) compared with unroofing and marsupialization (UM) as primary intervention for chronic pilonidal disease. A retrospective review of a single institution experience over a 5-year period was performed. Sixty-eight patients were treated for pilonidal disease. Their mean age was 25 years. There were 42 males (62%) and 26 females (38%). Twenty-six patients underwent WLE and 42 had UM. The type of operation was based on surgeon preference, but individual surgeons were consistent with their choice. Mean time for final healing was significantly higher for WLE compared with UM (21 vs 6 weeks, P < 0.01). Postoperative complications and reoperative rate were significantly higher in the WLE group (P < 0.01). Nine patients in the WLE group (35%) required one or more reoperations compared with only one (2%) in the UM group (P < 0.01). This study demonstrates that WLE carries a high complication and reoperative rate. UM should be considered as the initial operation for most patients with pilonidal disease.

Although neutropenia is recognized as a risk factor for infection and compromised wound healing, there are little data regarding the specific impact of neutropenia on morbidity and mortality after placement of implanted central venous catheters (CVC). We conducted a retrospective review of children with a diagnosis of acute lymphocytic leukemia or aplastic anemia who received a CVC over a 5-year period. The absolute neutrophil count immediately before catheter placement was recorded. Three hundred eight catheters were placed in 195 patients with acute lymphocytic leukemia and 15 with aplastic anemia. Absolute neutrophil count was less than 0.5 x 10⁹/L in 105 cases (Group 1). The incidence of CVC removal for all causes and for infection within 100 days in Group 1 was 17.1 per cent and 11.4 per cent, respectively, compared with 7.9 per cent (P = 0.01) and 1.5 per cent (P < 0.0001) with absolute neutrophil count 0.5 x 10⁹/L or greater (Group 2). Infections included two cases of mucormycosis with one death. Ports were more likely than Hickman catheters (C. R. Bard Inc., Murray Hill, NJ) to be removed for all causes (P = 0.01) and for infection (P = 0.04). The placement of implanted central venous catheters in neutropenic children was associated with substantial infectious morbidity and one death. When possible, CVC, particularly ports, should be avoided in the presence of neutropenia.