Traditionally in pediatric oncology, the surgeon was in charge of performing the biopsy for pathologic evaluation. Open biopsy is thought to be the safest approach, obtaining adequate tissue for diagnosis, as well as for tissue markers and genetic analysis. However, the disadvantages of surgical biopsy are not trivial, starting from the time lag between the need for biopsy and the availability of the operating room, through complications of general anesthesia and abdominal or chest surgery, the recovery after the operation in terms of pain and length of hospitalization, and the added technical difficulties in the definitive surgery.
For more than a decade, we have an alternative for percutaneous imaging-guided biopsies.1,2 In the published series, the procedure was shown to be safe and accurate for liver biopsies, solid tumors, anterior mediastinal tumors, musculoskeletal malignances, lymphoma, and leukemia in the pediatric population.3–9 Percutaneous imaging-guided biopsy has the potential of becoming the leading procedure, as more and more data are collected with respect to high accuracy rates and low complication rates.
In our department, ultrasound (US)-guided core biopsy is the first choice whenever there is a need for tissue diagnosis in the pediatric population, and other means for biopsy including incisional biopsy are saved for special cases that are inaccessible percutaneously. Most of the procedures are performed under deep sedation by an anesthesiologist. We retrospectively reviewed our experience in the procedure in the pediatric population and assessed the accuracy rate, the safety, and the availability of the procedure.
MATERIALS AND METHODS
We retrospectively studied US-guided biopsies on pediatric population performed in our hospital between November 2003 and November 2011. This work was approved by our local Institutional Review Board.
The charts of these patients were reviewed retrospectively. Data collection included demographics, clinical data, including reasoning for the biopsy, time lag between the request for a biopsy and the procedure, sonographer, and anesthesiologist reports of the procedure, pathologic reports, and clinical follow-up immediately after the procedure. Long-term follow-up included clinical evaluation, imaging, and pathology reports taken from the definitive operations.
The biopsies were performed under sonographic guidance in the US suite. The team included a radiologist (pediatric radiologists or senior staff members of the sonography unit), US technician, a nurse, and an anesthesiologist, in all cases carried out under deep sedation. Patients were never intubated. We use sonographic-guided biopsies, whenever feasible, from enlarged lymph nodes through anterior and posterior mediastinal mass, chest wall, abdominal solid organs, as well as peritoneal and omental infiltration and lytic bone lesions (including skull). Intra-abdominal and mediastinal biopsy cases were admitted for overnight hospitalization, whereas most of the other procedure were carried out as an outpatient procedure.
Before the procedure, the parents were informed of the need for the biopsy, the steps of the procedure, and the benefits and risks of the procedure, and they signed an informed consent. Anesthesiologist informed the parents regarding the need for deep sedation, and they signed an informed consent for sedation as well.
After a thorough examination of the child, the site for biopsy, the positioning of the child, and the probe were chosen. In musculoskeletal biopsies, the biopsy site and tract were chosen in coordination with the orthopedic surgeon in charge of the future definitive surgery and marked in the chart of the patient. Then sedation was given by the anesthesiologist, and preparation for the procedure including sterilization and covering was done by the radiologist. For local anesthesia, we used 1% lidocaine mixed with bicarbonate in a 10 to 1 ratio. In cases of deep sedation, a supplementation of oxygen by mask and continuous monitoring of blood pressure, pulse, and oxygen saturation were mandatory.
Using the same sedation, at the end of biopsy, a bone marrow biopsy was performed by the oncologists, as required.
For the core biopsy, we used a spring-loaded core biopsy needle (Super Core Biopsy Instrument; Angiotech), 14- to 18-G caliber. The 14-G caliber was mostly used in musculoskeletal biopsies, and for liver biopsies we usually used 18-G needle.
Core biopsies were sent for pathologic evaluation (3 cores in most cases), molecular biology, and genetic studies as needed.
All positive and negative results were defined as true positive or negative and false positive or negative on the basis of pathologic evaluation of the biopsy and the subsequent pathology diagnosis obtained from any other procedure (additional biopsy, incisional biopsy, or resection of tumor). Thus, false-negative result was defined as a failure of the percutaneous biopsy to identify tumor at the biopsied site, as was identified by other means.
We assessed the availability of the biopsy defined as the time lag between the request for biopsy and the procedure itself, immediate and delayed complications related to the biopsy, and the sedation, sensitivity, specificity, and accuracy rate of the procedure.
In cases of false-negative result, we studied the reasons for failure and the additional means taken to reach diagnosis.
Altogether we had 213 biopsies performed on 191 patients; in 17 of them, there was no question of malignancy and we exclude them from the study. Of the remaining 196 biopsies, 40 were performed on known oncologic patients suspected of having tumor progression (8) or tumor recurrence (32), and in 156 cases the biopsy was performed to establish diagnosis. The patient demographic can be found in Table 1.
Most of the patients were referred by pediatric oncologist (83%) and the minority by pediatric infectiologist and hepatologist. The suspicion of malignancy and indications for biopsy were clinical in 51 cases and imaging based in 145 cases: in 97 cases, a single modality (25 US, 46 computed tomography [CT], 13 positron emission tomography, 13 magnetic resonance imaging) and in 48 cases a combination of imaging modalities. In cases where the positron emission tomography CT raised a suspicion of tumor progression or recurrence, the site was chosen after consultation with the nuclear medicine team.
The mean time lag between the request for a biopsy and the procedure was 2 days (between the same day and 19 d). Seventy-six patients were biopsied on the same day of the request for biopsy (38%), and 53 patients underwent a biopsy on the next day (26.9%); thus, around 65% of the patients underwent a biopsy within a day.
Core biopsy needle caliber was 18 G in 50% of cases, 16 G in 39.7%, and 14 G in only 0.08%. The mean passes were 3.8, ranging from 1 to 7.
Sites of biopsy were mainly the neck, mediastinum, liver, and abdomen (both intraperitoneal and retroperitoneal). The sites of biopsy are shown in Table 2.
We performed 138 biopsies on patients with tumors at the biopsy site, of which 4 were missed by the initial biopsy and regarded as false negative. A total of 134 biopsies diagnosed correctly the tumor at the biopsy site and were read as true positive. In 58 biopsies, no malignancy was found, and follow-up was negative for malignancy, and thus they were read as true negative. We had no false-positive case. Repeated core biopsy was required in 7 cases. In 4 of them, the initial biopsy missed a tumor, and, in the other 3 biopsies, the additional biopsy was requested by the referring physician, because of lack of confidence in the initial diagnosis, and, in all of them, the additional biopsy confirmed the diagnosis in the first biopsy. The additional biopsies were percutaneous in 4 cases (3 under US and 1 under CT guidance) and in 3 cases an incisional biopsy was performed (2 in the mediastinum and 1 pelvic lymphadenopathy).
The final diagnosis of our patients is described in Table 3. The sensitivity of our US-guided core biopsy was 97.1%, the specificity was 100%, and the accuracy was 97.9%.
We had 2 cases of major complications related to the biopsy: bleeding from biopsy site in one patient and tumor seeding in another patient. The first patient was a 2-year-old girl with acute myeloid leukemia who developed liver failure and signs of liver fibrosis in sonography. She needed emergent liver biopsy for pathologic differentiation of the primary disease and fibrosis. Under maximal treatment, her coagulation factors could not be normalized. Her general condition was too poor to survive an incisional biopsy, and the percutaneous biopsy was considered a salvage procedure. The parents were involved in the decision to perform the biopsy and were informed of the added risk of bleeding from the biopsy site. She underwent US-guided liver needle biopsy in the intensive care unit. We took 2 cores of 18 G. After the procedure, she bled, reducing hemoglobin level by 2 g% and stabilized after 1 U of packed cells and additional coagulation factors.
The second patient underwent a biopsy of a kidney lesion, a Wilms tumor. The patient was a 6-year-old kid with flank pain and sharp decrease in pressure. On imaging studies, a kidney lesion was noticed. It was a multicystic lesion with debris and internal bleeding, and because there was a differential diagnosis (necrotic Wilms tumor or a traumatic lesion), a biopsy was taken percutaneously. Nine months later, on follow-up, she developed a local recurrence in the tract of the biopsy. The lesion was resected and the treatment protocol was adjusted as required. On long-term follow-up (up to 5.5 y), she is disease free.
Deep sedation was given in 148 cases; we found no complication related to sedation.
It is not uncommon for pediatric patients to require a tissue diagnosis. Many of them will soon after be diagnosed as oncologic patients and started on treatment; a few others will be diagnosed of having a chronic infection; and the rest will be diagnosed of having a reactive lymphadenopathy and will be discontinued from follow-up.
A diagnostic procedure that is prompt, accurate, and with the least number of complications is the procedure of choice. Percutaneous imaging-guided biopsy has the potential of becoming the leading procedure, as more and more data are collected regarding high accuracy rates and low complication rates.
The traditional way for achieving tissue diagnosis was an open biopsy in the operating room. Along with the advantages of having a “good bite” for the pathologists, there is the down side of having an operation under general anesthesia, with comorbidities related to anesthesia and laparotomies or thoracotomies and prolonged hospitalization with its high costs. There is also a time lag between the clinical need for biopsy and the operating room availability for the procedure, delaying the needed treatment and putting additional stress on the patients and their families.
Core needle biopsy presents several advantages compared with an open, surgical biopsy. The procedure requires conscious sedation instead of general anesthesia with intubation that can be done with monitoring in the US suite instead of the operating room, and our series had few complications. Avoiding a major incision also avoids potential surgical complications such as wound infection, wound dehiscence, and even abdominal compartment syndrome.10 Analgesic requirements after biopsy are minimal, if at all, and the children in our series can undergo their biopsy as an outpatient and return home to await the results of their biopsy.
Our accuracy rate was very high (97.9%), as good as or even higher compared with most of the published series.3,6–9,11 Of the 4 cases, we missed on the first biopsy, 3 were secondary to radiologist error in choosing the best sonographic window to sample the suspected site, and soon after a second US-guided biopsy corrected the technical error. In the other cases, the site was a small lesion in the mediastinum and the percutaneous biopsy was difficult. The radiologist took 3 cores in an 18-G biopsy gun. The repeated biopsy was incisional.
The use of core biopsy in the diagnosis of neuroblastoma has not yet been widely accepted. Thiesse et al12 used fine-needle aspiration and diagnosed correctly neuroblastoma in 92% of cases. The cytology was inadequate for the additional test, and N-MYC analysis was available in only 72%. There were no complications. Gupta et al13 compared incisional with percutaneous biopsy in neroblastoma. In his study, insufficient tissue occurred in both groups, more frequently in the percutaneous biopsy group. Major complications occurred in both groups, with major bleeding in 4/11 patients in the percutaneous biopsy group and 3/13 in the incisional biopsy group. There were additional postoperative complications in 2 patients—duodenal enterotomy and postoperative small bowel obstruction. We performed core biopsy in 35 patients with neuroblastoma (31 neuroblastoma and 4 ganglioneuroma). In all patients, both the diagnosis and N-MYC status were obtained. There were no complications or need for repeat biopsy in these patients. In no case of neuroblastoma in our series was our treatment ultimately changed on the basis of a lack of tissue or tumor markers.
There was one major complication of seeding of tumor cells at the biopsy site (0.007% in our series) in a renal biopsy. It is a documented complication, in 0.009% to 0.01% in recent review.14
Bleeding from biopsy site was seen in 1 patient, and, as detailed in the Results section, it was a special case of salvage biopsy under suboptimal conditions. As noted in the Results section, the patient stabilized after additional coagulation factors and required 1 U of packed cells. In a study by Oshrine and colleagues, similar population of patients underwent liver biopsies—either transjugular or percutaneous. They concluded that in this group of patients the complication rates (mainly bleeding) are high.15 Our rate of major complication was <1%.
The majority of our patients (65%) were biopsied within a day from the request for a biopsy. This high availability could be achieved mainly because no operating room was required. The availability of the US unit is higher, and there is more flexibility, because the routine studies are relatively short and an additional procedure can be easily added in between. There is a team work, with close collaboration between the oncologists and the pediatric radiologist, working together to optimize the care of the oncologic patients.
To conclude, we find US-guided core biopsy for suspected malignancy in the pediatric population to be highly available, safe, and very accurate. Our findings support that of others who found the procedure to be safe and quite accurate. Being a minimally invasive procedure, there is less morbidity, hospitalization length, and reduced costs compared with the alternative incisional biopsy in the operating room.
Saving additional tissue in the bank is not a routine procedure in our hospital. If the need will be raised by the oncologists, in the future we can add in most cases additional cores for this purpose.
The variability of the biopsy sites and tumors in our series reflects the real population in clinical life. It serves to show our use of US-guided core biopsy as the first choice for diagnoses in variable different cases.
1. Hoffer FA .Interventional radiology in pediatric oncology.Eur J Radiol. 2005; 53:3–13.
2. Hogan MJ, Hoffer FA .Biopsy and drainage techniques in children.Tech Vasc Interv Radiol. 2010; 13:206–213.
3. Mc Crone L, Alexander S, Karsill C, et al .US-guided percutaneous needle biopsy of anterior mediatinal masses in children.Pediatr Radiol. 2012; 42:40–49.
4. Amara JG, Schwartz J, Chait P, et al .Sonographically guided percutaneous liver biopsy in infants: a retrospective review.Am J Roentgenol. 2006; 187:644–649.
5. Pietrobattista A, Fruwirtg R, Natali G, et al .Is juvenile liver biopsy unsafe? Putting an end to a common misapprehension.Pediatr Radiol. 2009; 39:959–961.
6. Garrett KM, Fuller CE, Santana VM, et al .Percutaneous biopsy of pediatric solid tumors.Cancer. 2005; 104:644–652.
7. Van Sonnenberg E, Wittich GR, Edwards DK, et al .Percutaneous diagnostic and therapeutic interventional radiologic procedures in children: experience in 100 patients.Radiology. 1987; 162:601–605.
8. Garrett KM, Hoffer FA, Behm FG, et al .Interventional radiology techniques for diagnosis of lymphoma or leukemia.Pediatr Radiol. 2002; 32:653–662.
9. Shin HJ, Amaral JG, Armstrong D, et al .Image-guided percutaneous biopsy of musculoskeletal lesions in children.Pediatr Radiol. 2007; 37:362–369.
10. Chung PH, Wong KK, Lan LC, et al .Abdominal compartment syndrome after open biopsy in a child with bilateral Wilms tumor.Hong Kong Med J. 2009; 15:136–138.
11. Hassan SF, Mathur S, Magliaro TJ, et al .Needle core vs open biopsy for diagnosis of intermediate- and high- risk neuroblastoma in children.J Pediatr Surg. 2012; 47:1261–1266.
12. Thiesse P, Hany MA, Combaret V, et al .Assessment of percutaneous fine needle aspiration cytology as a technique to provide diagnostic and prognostic information in neuroblastoma.Eur J Cancer. 2000; 36:1544–1551.
13. Gupta A, Kumar A, Walters S, et al .Analysis of neeldle versus open biopsy for the diagnosis of advanced stage pediatric neronlastoma.Pediatr Blood Cancer. 2006; 47:875–879.
14. Robertson EG, Baxter G .Tumor seeding following percutaneous needle biopsy: the real story.Clin Radiol. 2011; 66:1007–1014.
15. Oshrine B, Lehmann LE, Duncan CN .Safety and utility of liver biopsy after pediatric hematopoietic stem cell transplantation.J Pediatr Hematol Oncol. 2011; 33:92–97.