Safety and Efficacy of Preoperative Portal Vein Embolization with Polyvinyl Alcohol in 58 Patients with Liver Metastases

OBJECTIVE

The objective of our study was to evaluate the safety and efficacy of transhepatic lobar portal vein embolization (PVE) using polyvinyl alcohol (PVA) particles to induce contralateral lobar hypertrophy in patients with liver-only metastases and normal underlying liver function.

MATERIALS AND METHODS

Fifty-eight consecutive patients with small predicted future liver remnants (FLRs) underwent PVE with PVA particles to induce hypertrophy of the contralateral hemi-liver before surgical resection of liver metastases. Total liver, right hemi-liver, and left hemi-liver volumes were calculated before and after embolization using a 3D workstation.

RESULTS

Eight patients underwent left PVE; 47, right PVE; and three, right and segment IV PVE. There were no major complications of the procedure. The mean increases in the ratio of the FLR to the total estimated liver volume after right, right and segment IV, and left PVE were 9%, 10%, and 3%, respectively; the corresponding mean hypertrophy ratios were 24.3%, 31.9%, and 1.5%, respectively.

CONCLUSION

Right PVE using PVA particles alone as the embolic agent is safe and effective in achieving left hemi-liver hypertrophy. In contrast, left PVE did not induce significant right hemi-liver hypertrophy in this patient population.

Upper urinary tract metastases from adenocarcinoma of the colon

An unusual presentation of colorectal metastasis to the upper urinary tract is reported. The metastasis manifested as a filling defect seen during antegrade pyelography. Cytologic evaluation of aspirated material demonstrated metastatic colonic adenocarcinoma. A dilated collecting system may be caused by intraluminal material including tumor and blood clots. Whenever fixed filling defects are encountered, urine cytology should be sent even in the absence of renal parenchymal involvement by tumor. The cytological evaluation may allow for prompt diagnosis and treatment.

Comparison of Survival Rates after Bland Arterial Embolization and Ablation Versus Surgical Resection for Treating Solitary Hepatocellular Carcinoma up to 7 cm

PURPOSE

The vast majority of hepatocellular carcinomas (HCC) occur in patients with underlying liver dysfunction, making surgical resection available to only a subset of patients with adequate hepatic reserve. This study analyzes the authors' results with bland arterial embolization combined with radiofrequency ablation (RFA) or percutaneous ethanol injection (PEIT) compared with surgical resection for the treatment of solitary HCC up to 7 cm in size.

MATERIALS AND METHODS

A retrospective review of all patients undergoing either surgical resection or bland embolization combined with local ablation for solitary HCC between January 1996 and August 2002 was performed. Progression-free survival rate and overall survival rate were calculated by the Kaplan-Meier method.

RESULTS

There were 40 patients who underwent surgical resection and 33 patients who underwent embolization and ablation. Age, gender, and size of the treated lesion were not significantly different between the groups. The embolization/ablation group had more patients classified as Okuda stage II (P<.001). The surgical group had a longer median recurrence-free survival rate (53.1 vs 25.1 months). With a median follow-up of 23 months, the 1-, 3- and 5-year actuarial overall survival rates were 97%, 77%, and 56% for the embolization/ablation group and 81%, 70%, and 58% for the surgical group, respectively. There was no statistical difference in overall survival rates (P=.20).

CONCLUSIONS

Bland arterial embolization in combination with ablation is effective in treating solitary HCC lesions up to 7 cm and achieves similar overall survival rates to surgical resection in selected patients.

Displacement of Endoscopically Placed Plastic Biliary Endoprostheses into the Duodenum with a Simple Transhepatic Technique

The authors to describe the technique and report the results of percutaneous displacement of endoscopically placed plastic biliary endoprostheses into the duodenum at the time of transhepatic intervention in 34 patients. Displacement into the duodenum was effected by simply passing a guide wire through the stent, and then pushing it into the gut with a catheter. Thirty-three of 36 stents (92%) were successfully displaced in this manner. No complications related to stent passage out of the gastrointestinal tract were encountered. Plastic biliary stents can be safely displaced into the duodenum when patients undergo transhepatic biliary procedures, thus avoiding repeated endoscopy.

CT–guided Transvenous or Transcaval Needle Biopsy of Pancreatic and Peripancreatic Lesions

PURPOSE

To evaluate the safety and efficacy of direct computed tomography (CT)-guided fine needle aspiration biopsy (FNAB) of pancreatic and peripancreatic masses via a posterior approach that traverses the inferior vena cava (IVC) or renal vein.

MATERIALS AND METHODS

From January 2000 to July 2003, 55 patients underwent 58 biopsies of masses located within the pancreas (n = 28) or in a peripancreatic location (n = 30) with use of a posterior approach that crossed the IVC or renal vein. Biopsies were performed with needles ranging in size from 18 to 22. Cytology reports and medical records of all patients were retrospectively reviewed to evaluate diagnostic accuracy and complication rates.

RESULTS

Masses were safely accessed with a direct (noncoaxial) pathway traversing the IVC (n = 54), renal vein (n = 4), or both (n = 3). Overall diagnostic accuracy was 86% (50 of 58). Cytologic examination was positive for malignancy in 39 of 58 biopsies (67%). Benign lesions were demonstrated in 12 of 58 biopsies. In seven cases (12%), the sample was deemed nondiagnostic. Of those, four were diagnosed later by endoscopy (n = 1), surgical biopsy (n = 2), or repeat FNAB (n = 1). A false-negative result was noted in one case, which was later diagnosed by repeat biopsy. A total of four inadvertent passes through the right renal artery were recorded. CT evidence of perilesional blood was seen in eight of 55 patients (eight of 58 cases). Three of these occurred after a passage via the right renal artery. All patients remained asymptomatic, and no transfusion or change in management was necessary. No other complications occurred.

CONCLUSION

CT-guided noncoaxial FNAB with an approach that traverses the IVC or renal vein is safe and effective in obtaining diagnostic specimens from pancreatic and peripancreatic masses.

Factors associated with pneumothorax and pneumothorax requiring treatment after percutaneous lung biopsy in 443 consecutive patients

PURPOSE

To describe patient- and procedure-related factors associated with post-biopsy pneumothorax and those that require intervention.

MATERIALS AND METHODS

Patient and procedure data from all lung biopsies performed at a single center between January 2000 and July 2001 were recorded prospectively. Data included patient demographics, lesion size, lesion depth from skin, needle size, number of passes, patient position during biopsy, imaging method used (computed tomography/fluoroscopy), if sedation was used, occurrence of pneumothorax and whether the pneumothorax required treatment. Patient charts were retrospectively reviewed to determine smoking history, as well as previous ipsilateral chest surgery or radiation therapy. Univariate and multivariate analysis was performed, and P <.05 was considered significant.

RESULTS

Four-hundred fifty-three biopsies were performed on 443 patients. One-hundred six patients (23.4%) had post-biopsy pneumothorax and 31 patients (6.8% overall, 29.2% of pneumothorax group) required intervention. By univariate analysis, increased patient age, smaller lesion size, increased depth from skin, supine position, and no history of surgery were significant predictors of biopsy-related pneumothorax. However, only increased patient age, supine position, no history of ipsilateral surgery, and history of smoking were associated with pneumothorax that required intervention. By multivariate analysis, increased patient age, smaller lesion size, and no history of surgery predicted pneumothorax; supine position, history of smoking, and no history of ipsilateral surgery predicted which patients with pneumothorax would require treatment.

CONCLUSION

Independent risk factors for pneumothorax include increased patient age, smaller lesion size, and no history of surgery. Previous surgery and prone positioning during biopsy appear to provide a "protective effect" against clinically significant post-biopsy pneumothorax.

Placement of subcutaneous chest wall ports ipsilateral to axillary lymph node dissection

Without scientific evidence, practice has been to avoid placing chest wall ports ipsilateral to axillary lymph node dissection. Because the lymphatic system in the chest wall drains via both the internal mammary and axillary nodes, it seems that this practice might unduly restrict venous access options. Our study was designed to evaluate outcome after placement of chest wall ports ipsilateral to axillary lymph node dissection in patients with breast cancer. Twenty-eight patients were studied in this retrospective two-institution review. The incidence of lymphedema after port placement (3.6%) was no higher than that reported after axillary lymph node dissection alone.

Pylephlebitis and liver abscess mimicking hepatocellular carcinoma

OBJECTIVE

To characterize the clinical and radiographic findings in patients with pylephlebitis and liver abscess with an emphasis on the findings that help to differentiate this disorder from portal vein occlusion associated with hepatocellular carcinoma.

METHODS

We analyzed the clinical findings and radiographic images of four patients with pylephlebitis and liver abscess(es) who had been misdiagnosed as having hepatocellular carcinoma with portal vein thrombosis. Their medical records were reviewed in terms of clinical presentation, physical findings, laboratory data, treatment, and follow up.

RESULTS

All patients undergoing color duplex ultrasonography had an echogenic thrombus within an expanded portal vein with negative color-flow findings within the thrombus. Contrast enhanced CT in all the patients demonstrated portal vein thrombosis associated with "liver masses". An intra-abdominal site of infection responsible for the subsequent ascending infection of the portal vein and liver was not identified in any patient on initial CT scan. At presentation, all patients were febrile and three of them had an elevated white blood cell count as well. All patients showed abnormalities of liver function.

CONCLUSIONS

Liver abscess(es) associated with pylephlebitis may mimic hepatocellular carcinoma with portal vein thrombosis. Clinical features that help to distinguish the two entities include presence or absence of fever, elevated white blood cell count, elevated alpha-fetoprotein, cirrhosis, and risk factors for hepatocellular carcinoma.

Variant hepatic arterial anatomy revisited: digital subtraction angiography performed in 600 patients

PURPOSE

To evaluate and describe the prevalence of hepatic arterial variants seen at digital subtraction angiography in a large series of patients.

MATERIALS AND METHODS

Data were collected prospectively by using an arterial anatomy database questionnaire that was completed at the time each visceral angiographic examination was performed from May 1996 to October 2000.

RESULTS

Six hundred patients underwent at least one visceral angiographic examination at one institution during the study period. Three hundred sixty-eight (61.3%) patients had the standard hepatic arterial anatomy. One hundred nineteen (19.8%) patients had variant left hepatic arteries (LHAs), and 89 (14.8%) had variant right hepatic arteries (RHAs). Twenty-eight (4.7%) patients had a variant anatomy involving both the LHA and the RHA. Twenty-four (4.0%) patients had a variant origin of the common hepatic artery (CHA) arising from either the superior mesenteric artery (SMA) or the aorta. In two patients, the proper hepatic artery (PHA) was the first branch of the SMA and the gastroduodenal artery (GDA) was a branch of the celiac axis. Double hepatic arteries were seen in 22 (3.7%) patients. Trifurcation or quadrifurcation of the GDA was seen in 50 (8.3%) patients, and the GDA originated distal to one hepatic artery in 25 (4.2%) patients in whom both hepatic arteries originated from the CHA.

CONCLUSION

A replaced LHA was less common than has been previously reported, and in two cases, the PHA arose from the SMA. Digital subtraction visceral angiographic results are comparable to results of seminal angiographic studies in which the cut-film technique was used.

Portal vein embolization with use of polyvinyl alcohol particles

Preoperative portal vein embolization has been used as method of inducing compensatory hypertrophy in the future remnant liver since it was first described in the late 1980s. Many different vascular embolic agents have been successfully used for this procedure, and there is no general consensus regarding which is the best agent. Polyvinyl alcohol (PVA) particles commonly used for arterial embolization come in many sizes, are readily available, and are easy to administer via conventional catheters. We describe an easy, safe, and effective method of the use of PVA particles for portal vein embolization.

Particle embolization of hepatic neuroendocrine metastases for control of pain and hormonal symptoms

PURPOSE

To evaluate treatment outcome with respect to the indication for treatment in patients with neuroendocrine tumors metastatic to the liver undergoing hepatic artery embolization with polyvinyl alcohol (PVA) particles.

MATERIALS AND METHODS

Charts and radiographs were reviewed of 35 patients undergoing 63 separate sessions of embolotherapy between January 1993 and July 1997. Patient demographics, tumor type, indication for embolization, and complications were recorded. Symptomatic and morphologic responses to therapy were noted, as well as duration of response.

RESULTS

Fourteen men and 21 women underwent embolization of 21 carcinoid and 14 islet cell tumors metastatic to the liver. These patients underwent 63 separate episodes of embolotherapy. Of 48 episodes that could be evaluated, response to treatment was noted following 46 episodes (96%). The duration of response was longest in patients treated for hormonal symptoms with (17.5 months) or without (16 months) pain, and was shortest (6.2 months) when the indication was pain alone. Complications occurred after 11 of the 63 embolizations (17%), including four (6%) deaths. Cumulative 5-year survival following embolotherapy was 54%.

CONCLUSION

Hepatic artery embolization with PVA particles is beneficial for patients with neuroendocrine tumors metastatic to the liver and may be used for control of pain as well as hormonal symptoms. This therapy should be used cautiously when more than 75% of the hepatic parenchyma is replaced by tumor.

Particle embolization for hepatocellular carcinoma

PURPOSE

To evaluate the outcome of all patients undergoing particle embolization for hepatocellular carcinoma at a single institution from January 1, 1993, through December 31, 1995.

MATERIALS AND METHODS

The charts and radiographs of all patients undergoing particle embolization during the study period were reviewed. The following information was collected: patient demographics, Child class and Okuda stage, number of embolization treatment sessions, length of hospital stay, complications related to the embolization procedure, including postembolization syndrome, current patient status, and date of death.

RESULTS

Forty-six patients underwent 86 embolization sessions during the study period. Postembolization syndrome developed after 70 of the 86 sessions (81%); in four cases (4.6%) this required treatment that extended the patient's hospital stay. Three other complications occurred (3.5%), including a splenic infarct and two episodes of transient hepatic failure, all treated supportively. There was one death within 30 days, but it was not directly attributable to embolotherapy. Follow-up was available for all of the patients who underwent treatment. Thirty-four patients were classified as Child class A, and 12 were classified as Child class B. Thirty patients were classified as Okuda stage I, 14 were classified as Okuda stage II, and two were classified as Okuda stage III. Overall actuarial survival was 50% at 1 year and 33% at 2 years. There was a statistically significant difference in survival between Okuda stage I and stage II patients, but not between Child class A and class B patients.

CONCLUSION

Particle embolization for hepatocellular carcinoma is well tolerated and demonstrates actuarial survival of 50% at 1 year and 33% at 2 years.

Clinical factors associated with positive bile cultures during primary percutaneous biliary drainage

PURPOSE

To evaluate the utility of routine bile cultures and to determine the risk factors for bacterial colonization of the bile as well as the biliary flora in patients with biliary obstruction undergoing primary percutaneous biliary drainage.

MATERIALS AND METHODS

Between October 1995 and January 1997, bile cultures were prospectively obtained in all patients undergoing percutaneous biliary drainage. Seventy-six patients underwent 86 procedures. Culture results were correlated with clinical, laboratory, and demographic variables. The antibiotic sensitivities of cultured organisms were examined.

RESULTS

Fever, previous endoscopic or percutaneous biliary instrumentation, and bilioenteric anastomosis were significant predictors of a positive bile culture. In the absence of any of these indicators, bile cultures were unlikely to be positive. Enterococcus species was the organism isolated most commonly. Yeast, gram-negative aerobic bacilli, and Streptococcus viridans followed in frequency.

CONCLUSION

Bile cultures provide valuable information that was useful for planning antibiotic prophylaxis and treatment. The likelihood of positive bile cultures can be predicted based on certain clinical variables. Continued investigation is needed to better predict bacterial flora in individual patients. Given the association between previous instrumentation and biliary colonization, noninvasive imaging modalities should be exhausted before invasive procedures are performed for solely diagnostic purposes in patients with biliary obstruction.

Outpatient treatment of iatrogenic pneumothorax after needle biopsy

PURPOSE

To evaluate prospectively the feasibility of treating iatrogenic pneumothorax after fine-needle aspiration biopsy (FNAB) of the lung with short-term placement of a small-caliber chest tube and to determine whether a 1-hour clamping trial is adequate to identify patients with persistent air leak.

MATERIALS AND METHODS

All patients undergoing FNAB of lung masses over a 28-month period were entered into the study. Patients with symptomatic, enlarging, or greater than 30% pneumothorax were treated with an 8-F chest tube. After 2 hours, the chest tube was clamped, and if the lung remained expanded for an additional hour, the chest tube was removed and the patients were discharged after a brief observation period. Patients were followed up by telephone after 24 hours.

RESULTS

Three hundred fifteen patients underwent FNAB of the lung. Sixty-eight patients (22%) developed a pneumothorax. Chest tubes were placed in 14 patients (4%): Six patients (2%) required admission to the hospital (four for air leaks), and the other eight patients were treated successfully as outpatients, with removal of the chest tube before discharge the day of FNAB.

CONCLUSION

Patients who develop clinically important pneumothorax after FNAB can be safely treated with short-term, small-caliber chest tubes and require hospital admission only if they demonstrate evidence of continued air leak.

The use of MRI in the diagnosis of occult hip fractures in the elderly: a preliminary review

Establishing the diagnosis of a non-displaced hip fracture in an elderly patient can be a prolonged and costly procedure, involving hospital admission, several days of bed rest, and a bone scan 3 to 5 days later. The authors evaluated 10 hips in 10 patients with a questionable diagnosis of non-displaced hip fracture. Magnetic resonance imaging (MRI) soon after admission revealed four patients with acute hip fractures who were subsequently treated. The other six patients, whose scans were negative for either femoral neck or intertrochanteric fractures, were mobilized. The authors show that, through the use of an immediate MRI study of a questionable hip fracture, the prolonged recumbency and inherent costs associated with awaiting a positive bone scan can be avoided.