Microsoft word - a review of complications.doc

A REVIEW OF COMPLICATIONS OF INTERVENTIONAL ULTRASOUND
PROCEDURES IN THE ABDOMEN

The Professional and Educational Standards Committee of EFSUMB committee plans to publish
a series of guidelines for ultrasound guided interventional procedures. We have already published evidence based guidelines on amniocenteses and chorionic villus sampling. As part of our work on such guidelines the committee publishes below a review of complications of interventional ultrasound guided procedures in the abdomen written by Dr Elisabetta Buscarini. At the end of this document you will find a few guidelines (needles and guidance, and precautions to be taken). We hope that our readers will find this publication useful in their clinical work and the committee thoroughly thanks Dr Buscarini for her contribution. Any comments on the review are welcomed and can be sent to EFSUMB’s General Secretary Mrs Gianna Stanford (e-mail: [email protected]).
Lil Valentin
EFSUMB’s Educational and Professional Standards Committee Chairperson

REVIEW OF INTERVENTIONAL ULTRASOUND IN THE ABDOMEN
“safety first”
Interventional ultrasound (US) includes invasive procedures carried out under US guidance for diagnosis and therapy. Diagnostic procedures are undertaken for cytology or tissue sampling, generally performed with a fine needle (FN), which has a calibre less than 1 mm. Therapeutic procedures performed under US guidance are for drainage of fluid collections, or of obstructed collecting systems (typically US guided nephrostomy), or of hollow organs for nutritional purpose. Tumour ablation either by injection of chemicals (mainly ethanol) or by deposition of thermal energy with radiofrequency electrode needles or laser fibers is another therapeutic area. The small calibre of biopsy needles and the safety offered by US guidance has made interventional US a minimally invasive procedure; nevertheless fatal and major complications are reported, generally caused by abdominal interventional procedures. Therefore, physicians performing interventional US should have a thorough knowledge of the relevant literature and of reported complications to improve their technical choices, to reduce the risk of complications, and to minimise the consequences of complications when they occur. It is important to learn how to prevent and how to correct complications.
Definition
One problem when discussing complications is to ensure that everyone is speaking about the same thing. Complications range from trivial incidents to major life-threatening crises, and without any generally agreed definitions it is difficult to make sense of A complication can be defined as an unfavourable event, unexpected even if predictable, occurring because of
the invasive procedure, in spite of technical accuracy of the procedure.
Depending on their clinical and biological impact complications are classified as: Minimal, when they cause transient inconvenience, they do not imply a significant worsening of the patient’s
condition, they resolve spontaneously or with minimal care, even if they may require a short period of intensive check- up of some parameters (e.g., blood pressure) Major, when they induce a significant worsening of the clinical condition of the patient and require substantial care
(e.g., blood transfusion, resuscitation, surgery) with delayed hospital discharge or renewed hospitalization
According to the timing of appearance complications are divided into early and late.

Complications of diagnostic US-guided procedures
Mortality rate due to abdominal biopsies ranges from 0.001% to 0.038% as described in major studies, including questionnaire surveys obtained by multicentre and single institution series (1-8, Table I). Mortality and major complication rates are greatest for hepatic and pancreatic biopsies. However, cases of death have been described after biopsy of every abdominal organ. Two deaths out of 6,261 biopsied patients (0.03%) occurred after hepatic biopsy. Both fatalities were due to massive haemorrhage. They occurred in one patient with hepatocellular carcinoma (HCC) out of 2,293 patients who underwent biopsy for the diagnosis of HCC, and in one patient with liver haemangioma out of 157 patients who underwent biopsy for the diagnosis of liver haemangioma (9). Another three cases of death after puncture of liver haemangioma have been observed in a series from a single institution (8). These data suggest a significant risk of bleeding after biopsy of liver haemangioma. Therefore diagnostic work-up of liver haemangioma should be completed whenever possible by using imaging techniques. The risk of bleeding is very important in case of biopsy of angiosarcoma. In spite of the rarity of this tumor, fatalities have been reported after biopsy of angiosarcoma either in the liver (4, 10) or in the adrenal gland (2). Fatal complications after pancreatic biopsy are mainly due to severe pancreatitis, after puncture of a normal gland in the wrong assumption of a pancreatic mass (11,12). The rate of major complications after ultrasound guided biopsy is shown in table I. In a multicentre survey of haemorrhagic complications after liver fine needle biopsy (FNB) the rate was 0.13%. It did not seem related either to the needle type (aspiration or cutting needle) or to blood clotting function. The risk of liver haemangioma puncture was confirmed, because in two cases out of 157 hemorrhagic complications occurred (9). Even if spleen biopsy is a commonly feared procedure the related series showed no fatalities, but major complication rate was as high as 1.3% (10). Table I – Deaths, major complications, and tumor seeding observed after diagnostic or diagnostic and therapeutic interventional procedures (series 3, 6 and 7) guided by ultrasound. An intriguing complication is tumor seeding, which implies the dragging of a critical number of tumour cells along the needle track, their deposition in a favorable microenvironment and subsequent tumour growth. The time elapsing between the procedure and tumour seeding generally corresponds to a few months, even if in some instances it is as long as two years or more. The incidence general y varies between 0.003% and 0.036% but the exact incidence is difficult to determine, because only in a proportion of the patients is follow up complete. Tumour seeding correlates with needle calibre, number of biopsy passes into the tumor, and location of the tumor. Seeding is probably easier after a puncture of superficial tumours while it seems to be independent of tumour histology (14). However a high incidence of tumour seeding after pancreas tumour biopsy is frequently reported, even if in one large series of pancreas biopsies no case of seeding occurred (15). Another analysis of 33 reported series of pancreatic biopsies, including 2533 patients altogether, revealed 1 (0.039%) case of seeding (16). According to some authors biopsy should be avoided in patients who are candidates for surgery to avoid the risk of tumour seeding. The high diagnostic accuracy of imaging techniques strongly supports this point of view. The matter is still debated, even though the trend is to reduce invasive procedures. Irregularities of the needle surface to improve needle US visualization (the so-called echo-marker) may potentially increase the seeding. An in-vitro study has shown these irregularities to induce greater cell dragging after biopsy. Therefore echo-marker should be avoided (17). Tumour seeding seems to rarely have a clinical impact, and it generally does not affect the patient outcome.
Complications of therapeutic US-guided procedures
Drainage of abdominal collections is followed by a variable number of major complications and deaths. In a large series (886 patients) of abdominal drainages no death related to the procedure was reported, but major complications were observed in 77 Complications of percutaneous ethanol injection (PEI) in the treatment of HCC have been studied in a multicentre series of 1066 patients (19). Mortality rate was 0.09% (one death due to haemoperitoneum). Major complication rate was 3.2%. The complications comprised haemorrhage (9 cases), pleural effusion, hepatic or portal vein thrombosis, hepatic infarct, and liver abscess. Forty cases of severe pain with interruption of the procedure were described but not included in the major complications. Tumour seeding along the needle track was observed in 7 patients (0.6%). In another paper (20) tumour dissemination was found in 4 out of 348 patients (1.1%). An emerging percutaneous therapeutic option used either for small HCC or for liver metastases is radio-frequency (RF) thermal ablation. There are two different technologies: the expandable needle electrode and the cooled tip needle electrode. Complications of the cooled system have been described in a multicentre report (21), while those of the expandable system are being reported in a series from a single institution (22). It is certainly interesting to compare adverse effects of PEI and RF thermal ablation (Table II). However an “a priori” definition of complication is missing in the paper on PEI (19), and the definition of major complications in the RF cooled system series (21) differs from that used in the RF expandable system series (22). Number of patients
Major complications
Major compl.
74 (6.9%)
50 (2.2%)
number (rate)

Table II – Complications after percutaneous ethanol injection (16), the radiofrequencies cooled system port (21) and the
radiofrequencies expandable system (22). The reported data confirm that RF thermal ablation can be considered at least as safe as PEI for the treatment of liver tumors.
Needles and guidance
a. Fine and large (>1 mm) needles (aspiration and cutting), catheters, needle-electrodes b. Ultrasound guidance is sometimes difficult and may be replaced by computed tomography (CT).
Comments
a. Experimental and multicentre studies on fine needles show no effect of needle calibre (23) nor between aspiration and cutting needles on procedure related bleeding , whereas the use of large needles has been shown to be associated with an increased complication rate when compared to fine needles (1-9). b. In most of the series reporting percutaneous biopsies or therapeutic procedures the guidance of choice has been US, even if some difficulties in targeting the lesion (i.e.because of meteorism or obesity) may indicate the need to use CT guidance. Literature survey does not indicate any connection between complication rate and type of guidance (16).
Precautions to be taken

a. Careful patient history, check coagulation tests

Comments
a. A detailed clinical history can sometimes reveal a haemostatic defect even in the presence of normal routine coagulation tests. Coagulation tests which should routinely be evaluated before a percutaneous procedure on deeply located organs include: prothrombin time, partial prothrombin time, platelet count. The following values are generally considered safe: prothrombin time >40 per cent, partial prothrombin time < 5 sec above the upper limit, platelet count > 50,000/mm3 (9). Discontinuation of aspirin and anticoagulants (possibly replace by heparin) is advisable before a biopsy, but it has to be weighed against the thrombotic risk (24). b. When deciding the needle track any interposed structure between the abdominal wall and the target lesion or parenchyma has to be carefully evaluated so as to avoid passage through main blood vessels, gallbladder, or colon (16). Systematic check of the needle track with Doppler US can easily identify interposed vessels or vascular lesions d. It has been shown that the diagnostic accuracy of cytology significantly increases if two passes are made instead of one, but that it does not increase if three or more passes are made. It is therefore recommended to check immediately the adequacy of every specimen by a rapid staining. This can reduce risks of complication by saving an e. Experience of the operator and number of performed procedures are certainly important factors affecting the complication rate of percutaneous biopsies (26). Personal opinion and experience suggests that a number of 50 annual liver biopsies is needed for biopsy to be safe.
Concluding remarks
An ultrasound guided diagnostic procedure, even if it is considered a minimally invasive one, should only be performed if it is judged to be of benefit to the patient. It should not be performed if it can be replaced by a less invasive procedure. An US guided therapeutic procedure should only be performed if it can be expected to give a result equal to or better than that obtainable by a more invasive procedure (e.g.,a surgical intervention). 1. Livraghi T, Damascelli B, Lombardi G, Spagnoli I (1983) Risk in fine needle abdominal biopsy. JCU 11: 77-81 2. Smith EH (1984) The hazard of fine needle aspiration biopsy. Ultrasound Med Biol 10: 629-634 3. Weiss H, Duntsch U, Weiss A (1988) Risiken der Feinnadelpunktion. Ergebnisse einer Umfrage in der BRD (DEGUM- 4. Smith EH (1991) Complications of percutaneous abdominal fine needle biopsy. Radiology 178: 253-258 5. Fornari G, Civardi G, Capanna L, Di Stasi M, Rossi S, Sbolli G, Buscarini L & The Cooperative Italian Study Group (1989) Scand J Gastroenterol 24: 949-955 6. Weiss H (1994) Komplikationen der Feinnadel Punktion. DEGUM Umfrage 2. Bildgebung Imaging 61 suppl 2: 25-28 7. Nolsoe C, Nielsen L, Torp-Pedersen S, Holm HH (1990) Major complications and deaths due to interventional ultrasonography: a review of 8,000 cases. JCU 18: 179-184 8. Livraghi T, Lazzaroni S, Civelli L, Marks M, Meloni F, Vettori C. (1997) Risk conditions and mortality rate of abdominal fine needle biopsy. J Intervent Radiol 10: 57-64 9. Di Stasi M, Buscarini L, Bolondi L (1995) Ultrasound guided fine-needle liver biopsy : a multicentre survey of preprocedure evaluation and complication rates. J Intervent Radiol 10: 43-48 10. Hertzanu Y, Peiser J, Zirkin H. (1990) Massive bleeding after fine needle aspiration of liver angiosarcoma. Gastrointest 11. Evans WK, Ho CS, Mc Loughlin MJ, Tao LC (1981). Fatal necrotizing pancreatitis following fine needle aspiration biopsy 12. Brandt KR, Charboneau JW, Stephens DH, Welch TJ, Goellner JR (1993) CT- and US-guided biopsy of the pancreas. 13. Di Stasi M, Buscarini L, Cavanna L, Rossi S, Buscarini E, Silva M (1996) Complications of ultrasound guided fine- needle biopsy of the spleen: report on 110 patients and review of the literature. J Intervent Radiol 11: 43-46 14. Buscarini L. (1998) Complications of abdominal interventional ultrasound: the dissemination risk. JMU 19: 149-152 15. Di Stasi M, Lencioni R, Solmi L, Magnolfi F, Caturelli E, De Sio I, Salmi A, Buscarini L (1998) Ultrasound guided fine needle biopsy of pancreatic masses; results of a multicenter study. Am J Gastroenterol 93 : 1329-1333 16. Buscarini E, Di Stasi M. Complications of abdominal interventional ultrasound, pp 24. Poletto Ed., Milan, 1996 17. Buscarini E, Foroni R, Rossi S, Di Stasi M, Silva M, Marinone G, Degli Antoni G, Buscarini L (1997) Fine needles with echo markers: increasing cell dragging during biopsy. Acta Cytol 41: 1246-1249 18. Civardi G, Di Candio G, Giorgio A, Goletti O, Ceraioli T, Filice C, Caremani M, Buscarini L (1998) Ultrasound guided percutaneous drainage of abdominal abscesses in the hands of the clinician: a multicentre Italian Study. Europ J 19. Di Stasi M, Buscarini L, Livraghi T, Giorgio A, Salmi A, De Sio I, Brunello F, Solmi L, Caturelli E, Magnolfi F, Caremani M, Filice C (1997) Percutaneous ethanol injection in the treatment of hepatocellular carcinoma. A multicenter survey of evaluation practices and complication rates. Scand J Gastroenterol 32: 1168-1173 20. Ishii H, Okada S, Okusaka T, Yoshimori M, Nakasuka H, Shimada K, Yamasaki S, Nakanishi Y, Sakamoto M (1998) Needle tract implantation of hepatocellular carcinoma after percutaneous ethanol injection Cancer 82: 1638-1642 21. Livraghi T. Solbiati L, Meloni F, Scott Gazelle G, Halpern EF, Goldberg SN (2003) Treatment of focal liver tumors with percutaneous radio-frequency ablation: complications encountered in a multicenter study. Radiology 226:441-451. 22. Buscarini E, Buscarini L. RF thermal ablation with expandable needle of focal liver malignancies: complication report. 23. Scott Gazelle G, Haaga JR, Rowland DY (1992) Effect of needle gauge, level of anticoagulation, and target organ on bleeding associated with aspiration biopsy. Radiology 183:509-513 24. Moulton JS, Moore PT (1993) Coaxial percutaneous biopsy technique with automated biopsy devices: value in improving accuracy and negative predictive value. Radiology 186:515-522 25. Civardi G, Fornari F, Cavanna L, Di Stasi M, Sbolli G, Buscarini L (1988) Value of rapid staining and assessment of US- guided fine needle aspiration biopsy. Acta Cytol 32:552-554 26. Froelich F, Lamy O, Fried M, Gonvers JJ (1993) Practice and complications of liver biopsy. Results of nationwide survey in Switzerland. Dig Dis Sci 38:1480-1484

Source: http://www.duds.dk/rekommandation/Review%20of%20Complications.pdf