Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/radiol.2481061874
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nair, R. T.
Right arrow Articles by Shankar, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nair, R. T.
Right arrow Articles by Shankar, S.
(Radiology 2008;248:303-311.)
© RSNA, 2008

Vascular and Interventional Radiology

Biochemical and Hematologic Alterations Following Percutaneous Cryoablation of Liver Tumors: Experience in 48 Procedures1

Rashmi T. Nair, MD, Stuart G. Silverman, MD, Kemal Tuncali, MD, Nancy A. Obuchowski, PhD, Eric vanSonnenberg, MD 2, and Sridhar Shankar, MD 3

1 From the Division of Abdominal Imaging and Intervention, Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, Mass (R.T.N., S.G.S., K.T., E.v.S., S.S.); and Section of Abdominal Imaging (R.T.N.) and Department of Quantitative Health Sciences (N.A.O.), Division of Radiology, Cleveland Clinic Foundation, 9500 Euclid Ave, Hb6, Cleveland OH 44195. Received November 2, 2006; revision requested January 9, 2007; revision received May 7; accepted June 19; final version accepted, January 30, 2008. Address correspondence to R.T.N. (e-mail: nairr2{at}ccf.org).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 IMPLICATIONS FOR PATIENT CARE
 References
 
Purpose: To retrospectively determine the frequency and severity of various abnormal laboratory test values following percutaneous cryoablation of liver tumors and to estimate the correlation between laboratory test values and tumor and ablation volumes.

Materials and Methods: This HIPAA-compliant study had institutional review board approval. Informed consent was waived. Biochemical and hematologic laboratory values from 48 procedures in 39 patients (18 men and 21 women; age range, 29–86 years) who underwent magnetic resonance (MR) imaging–guided percutaneous cryoablation of 65 liver tumors (62 metastases, three hepatocellular carcinomas) were retrospectively reviewed. Changes in laboratory values at baseline and 0–6 hours and 1–2 weeks after the procedure were analyzed with respect to tumor and ablative margin volumes by using generalized estimating equations. Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels were correlated with percent maximal decrease in platelet count.

Results: Mean ablation zone volume was 67.3 cm3 ± 41.2 (standard deviation) (range, 7.3–191.4 cm3). AST and ALT values increased after all procedures and peaked at 6 hours (median change in AST value, +835 U/L; median change in ALT value, +614.5 U/L). Platelet count decreased after 47 procedures (mean maximal decrease, 92.3 x 109/L [38%]), reaching a nadir at 12–24 hours after 24 procedures (50%) and returning to normal in 31 (84%) of 37 procedures at 1–2 weeks. One procedure was complicated by disseminated intravascular coagulation that necessitated transfusion and arterial embolization. Myoglobin values increased after 21 (44%) of 48 procedures and peaked at 6 hours (trimmed-mean value, 183.4 µg/L). Ablative margin volumes were predictive of changes at 0–6 hours in AST (P = .02), ALT (P = .003), and myoglobin (P < .001) values. Percent maximal decrease in platelet count correlated with peak change in AST (r = 0.72) (P < .001).

Conclusion: Following percutaneous cryoablation of liver tumors, alterations in liver enzymes, myoglobin, and platelet count are common, are usually self-limited, and correlate with ablative margin volume—except for changes in platelet count, which correlate with changes in AST and ALT.

© RSNA, 2008


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 IMPLICATIONS FOR PATIENT CARE
 References
 
Cryoablation of liver tumors has been performed during open surgery or laparoscopically, by using ultrasonographic guidance, since the late 1960s (1). Although hepatic cryosurgery has overall low morbidity (13.8%) and mortality (1.6%) rates (1), biochemical and hematologic alterations occur. Severe thrombocytopenia, liver failure, and cryoshock (a syndrome of multiorgan failure and disseminated intravascular coagulation [DIC]) are occasionally seen (1–10).

Cryoablation of liver tumors can now be performed percutaneously, and data regarding its safety and efficacy are accumulating (11–14). However, few data are available regarding biochemical and hematologic alterations following percutaneous ablation (12,13). Thus, the purpose of our study was to retrospectively determine the frequency and severity of various abnormal laboratory test values following percutaneous cryoablation of liver tumors and to estimate the correlation between laboratory test values and tumor and ablation volumes.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 IMPLICATIONS FOR PATIENT CARE
 References
 
There was no industry support for this study. One author (S.G.S.) is a consultant for Galil Medical (Yokneam, Israel) but did not exercise control over data collected and information submitted for publication.

Patients
Approval of the institutional review board of Brigham and Women’s Hospital was obtained for our retrospective study of patients who were treated with magnetic resonance (MR) imaging–guided percutaneous cryoablation of liver tumors from January 1999 to September 2004. The study was in compliance with the Health Insurance Portability and Accountability Act. Informed consent was waived. Forty-eight consecutive procedures were performed in 39 patients. Among the 39 patients, 18 were men and 21 were women, and the age range was 29–86 years (mean age, 60.8 years).

A total of 65 liver tumors (62 metastases, three hepatocellular carcinomas) were treated during 48 procedures in 39 patients (Table 1). In 36 procedures, one tumor was treated; in eight procedures, two tumors were treated; in three procedures, three tumors were treated; and in one procedure, four tumors were treated. Mean pretreatment tumor size was 2.7 cm (range, 1–5 cm); mean tumor volume was 21.3 cm3 (range, 1.8–81.2 cm3).


View this table:
[in this window]
[in a new window]

  		Table 1. Liver Tumors Treated with Percutaneous Cryoablation in 39 Patients

 
Procedures
Percutaneous cryotherapy was performed by using a U.S. Food and Drug Administration–approved argon-based cryotherapy delivery system (Cryohit; Galil Medical) and was guided by using a 0.5-T open-configuration MR imaging system (Signa SP; GE Medical Systems, Milwaukee, Wis) with previously described methods (13). The procedures were performed by abdominal interventional radiology staff members (S.G.S., K.T., and E.v.S., with 10, 5, and 20 years of experience, respectively) and abdominal radiology fellows (R.T.N. and S.S., with 2 and 3 years of experience, respectively). Multiple MR imaging–compatible cryoprobes were placed in each tumor (range, one to five; mean number, 2.5; probe diameter, 2.1–2.4 mm). Two 15-minute freezes separated by a 10-minute thaw were executed at each probe station. All procedures were performed with the patients receiving general anesthetic, as is the practice in our institution. One gram of cefazolin sodium (Ancef; SmithKline Beecham Pharmaceuticals, Philadelphia, Pa) was administered intravenously before each procedure and every 8 hours thereafter for a total of three doses after the procedure.

In the first 12 procedures, to prevent myoglobinuria-induced acute tubular necrosis, 0.3 g d-mannitol (Abbott Laboratories, North Chicago, Ill) per kilogram of body weight was administered at the beginning of each procedure, followed by three ampules of sodium bicarbonate (50 mEq per ampule) in 5% dextrose in water at 150 mL per hour for 24 hours. For the remaining 36 procedures, d-mannitol and sodium bicarbonate were administered at the same doses only if postprocedural serum myoglobin levels increased above 1000 µg/L, as we did not observe marked myoglobinemia in the first 12 procedures.

All patients underwent MR imaging both before and within 24–48 hours after the procedure. All volume measurements were made on postprocedural MR images, as previous work has shown that intraprocedural ice balls correlate well with postprocedural MR imaging estimates of cryonecrosis (13). Coronal locator T2-weighted single-shot fast spin-echo (repetition time msec/echo time msec, 1160/180; section thickness, 8 mm; field of view, 42 cm), transverse T2-weighted fast spin-echo (5100/100; echo train length, 23; section thickness, 5 mm; field of view, 30 cm), axial T2-weighted single-shot fast spin-echo (1160/180; section thickness, 5 mm; field of view, 34 cm), and axial breath-hold T1-weighted dual-echo fast spoiled gradient-echo (150/2.2, 4.4; section thickness, 5 mm; field of view, 34 cm) images were obtained, along with transverse three-dimensional fast spoiled gradient-echo Fast Acquisition with Multiphase Efgre3D (FAME) (4.4/1.6; flip angle, 75°; section thickness, 5 mm; field of view, 34 cm; fat suppression) images before and after the intravenous injection of 20 mL gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ). These images were then reformatted in the sagittal and coronal planes.

Laboratory Tests
Serum aspartate aminotransferase (AST), alanine aminotransferase (ALT), bilirubin, alkaline phosphatase (ALKP), myoglobin, and creatinine levels, as well as platelet count, white blood cell (WBC) count, and hematocrit values, were recorded for all 48 procedures at baseline (within 30 days before the procedure) and 0–6 hours, 6–12 hours, 12–24 hours, and 1–2 weeks after the procedure (R.T.N., S.S.). Elevations in serum AST were classified as mild, moderate, or severe (normal range, 11–32 U/L; mild elevation, less than three times normal; moderate elevation, three to 20 times normal; severe elevation, more than 20 times normal). Similarly, elevations in serum ALT were classified as mild, moderate, and severe (normal range, 3–30 U/L; mild elevation, less than three times normal; moderate elevation, three to 20 times normal; severe elevation, more than 20 times normal). Elevations in myoglobin values were also classified as mild, moderate, and severe (normal range, 0–100 µg/L; mild elevation, 101–399 µg/L; moderate elevation, 400–1000 µg/L; severe elevation, >1000 µg/L). Changes in serum ALKP were classified as mild, moderate, and severe (normal range, 35–105 U/L; mild elevation, less than two times normal; moderate elevation, two to five times normal; severe elevation, more than five times normal). Changes in serum bilirubin and creatinine levels, platelet count, and hematocrit were recorded without classification. A platelet count of less than 100 x 109/L was considered to constitute clinically important thrombocytopenia. The WBC count was recorded as normal (4.0–11.0 x 109/L) or abnormal (<4.0 or >11.0 x 109/L).

Common laboratory abnormalities at baseline were anemia (23 [48%] of 48 procedures), mild to moderate elevation in ALT (16 [33%] of 48 procedures), elevated ALKP (12 [25%] of 48 procedures), abnormal WBC count (10 [21%] of 48 procedures), and mildly elevated AST (eight [17%] of 48 procedures). These abnormalities were thought to be related to either previous chemotherapy or underlying malignancy. Length of hospital stay and complications related to the procedure that were evident clinically or that resulted in further diagnostic or therapeutic interventions or additional observation (major complications were as defined in Society of Interventional Radiology practice guidelines) and the underlying reasons for these complications were reviewed.

Tumor and Ablation Volumes
Largest diameters of the tumors and ablation zones (new areas of nonenhancement) were measured in three orthogonal planes on MR imaging studies obtained before and 24 hours after the procedure, respectively. All measurements were performed by one author (R.T.N., with 2 years of experience). Volumes were calculated by using the formula for an ellipsoid (4/3 · {pi} · r1 · r2 · r3, where r [radius] is equal to half of the measured diameter in each of the three orthogonal planes). Ablative margin volume (which consists of ablated nontumoral liver and was defined as ablation zone volume minus tumor volume) was calculated in cases where the ablation volume extended beyond the tumor (56 of 65 tumors) (Fig 1). In tumors where ablation was incomplete (ie, the ablation zone did not encompass the tumor completely), the 24-hour postprocedural MR imaging study was reviewed to assess the extent of ablated normal liver parenchyma. In all nine tumors that were incompletely ablated, the ablation zone was confined within the tumor margins and no normal surrounding liver parenchyma was ablated (Fig 2).


Figure 1
View larger version (128K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1: Image in 54-year-old woman with colorectal carcinoma refractory to chemotherapy and hepatic metastasis. She was not a surgical candidate and was referred for ablation therapy. Axial contrast-enhanced T1-weighted three-dimensional fast spoiled gradient-echo FAME (4.4/1.6; flip angle, 75°; section thickness, 5 mm; field of view, 34 cm; fat suppression) image obtained on day 1 after ablation shows ablation zone volume (long arrows), tumor volume (short arrows), and margin of nontumoral ablated liver (ablative margin volume) (L).

 

Figure 2
View larger version (78K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2: Image in 68-year-old woman with non–small cell carcinoma of lung refractory to chemotherapy and metastasis to liver. The patient was not a surgical candidate and was referred for ablation therapy. Axial contrast-enhanced T1-weighted three-dimensional fast spoiled gradient-echo FAME image obtained with the same imaging parameters as Figure 1 on day 1 after ablation shows ablation zone volume (long arrows) and tumor volume (short arrows). Note that there is no ablation of surrounding normal liver.

 
Statistical Analysis
The mean for all laboratory test values for the entire group at each time point was calculated. The change in laboratory test values for the entire group relative to baseline values was calculated for each time period. The largest change from baseline for each laboratory test value ({Delta}max) for each patient, and the time interval in which it occurred, were recorded. The mean of the {Delta}max was calculated for the entire group. In addition, the {Delta}max for platelet count was expressed as a percentage of the baseline platelet count to standardize the observed change because there was wide variation among baseline platelet counts. The mean of the {Delta}max expressed as a percentage of the baseline platelet count was calculated for the entire group ({Delta}max percent platelet count). The mean ablative margin volume of the three procedures that resulted in marked thrombocytopenia (<100 x 109/L) was compared with the mean ablative margin volume of the remaining 45 procedures by using a two-tailed unpaired t test. Of 2160 total observations, 393 (18%) were missing and were therefore not included in the analysis.

The degree of correlation between maximal changes in serum AST, ALT, and {Delta}max percent platelet count was estimated by using the Pearson correlation coefficient.

We tested two null hypotheses: (a) biochemical and hematologic laboratory values at 6 hours and 2 weeks did not change from baseline and (b) any change in the biochemical and hematologic laboratory values from baseline was not related to tumor volume, ablative margin volume, or total ablation volume. The outcome variable was the change in the laboratory value from baseline to 6 hours and from baseline to 2 weeks.

We used a Wilks-Shapiro test to determine if changes in the biochemical and hematologic laboratory variables followed a normal distribution. For variables that were not normally distributed, we applied various transformations to normality. If a suitable transformation could not be found, we performed analyses on the basis of ranks of the outcome variable.

We fitted a generalized linear model to the data by using generalized estimating equations to handle clustered data (ie, multiple procedures in some patients). We specified a normal distribution with identity link and an exchangeable working correlation matrix. The value of the outcome variable at baseline was included as a covariate in all models. The predictor variables of interest were tumor volume, ablative margin volume, and total ablation volume. Ablative margin volume and total ablation volume were highly correlated (r = 0.89, P < .001). Tumor volume and total ablation volume were mildly correlated (r = 0.31, P = .035). Thus, to assess the simultaneous effect of these predictors on the outcome variables, we included in the model only tumor volume and ablative margin volume.

Software (SAS; SAS Institute, Cary, NC) was used for the statistical analysis.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
 IMPLICATIONS FOR PATIENT CARE
 References
 
Biochemical Laboratory Tests and Complications
All laboratory values reported refer to all 48 procedures unless otherwise specified. The mean AST value at baseline was 25.5 U/L ± 12.6 (standard deviation). AST levels increased significantly at 0–6 hours (P < .001; median change, +835 U/L) and peaked within 6 hours after the procedure in 45 (94%) of 48 procedures (Fig 3); the 0–6-hour values were not available for the remaining three procedures. At 1–2 weeks after the procedure, AST values returned to normal for 22 (52%) of 42 procedures. AST values remained mildly elevated at 1–2 weeks for 19 (45%) of 42 procedures and moderately elevated for one (2%) of 42 procedures. There was a small but statistically significant difference (P < .001; median change, +5 U/L) between baseline and 1–2 week values.


Figure 3
View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3: Line plot shows change in mean serum AST level with time. Bar graph depicts distribution of AST elevation according to severity at each time interval studied. Note: Normal AST = 11–32 U/L; mild elevation in AST = less than three times normal; moderate elevation in AST = three to 20 times normal; severe elevation in AST = greater than 20 times normal. Hash marks on x-axis indicate switch from days to weeks.

 
Mean ALT value at baseline was 29.1 U/L ± 19.7 and increased significantly at 0–6 hours (P < .001; median change, +614.5 U/L) (Fig 4). Peak values were most commonly seen at the 0–6-hour interval in 42 (89%) of 47 procedures (Fig 4). In four (9%) of 47 procedures, peak values were seen at the 6–12-hour interval. At 1–2 weeks after the procedure, ALT values returned to normal for five (10%) of 48 procedures and remained mildly elevated for 22 (46%) and moderately elevated for 11 (23%) procedures. The mean ALT value at 1–2 weeks was significantly different from that at baseline (P < .001; median change, +40.5 U/L). In one procedure, AST increased to 1225 U/L and ALT increased to 862 U/L, prompting observation in the hospital for one additional day.


Figure 4
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4: Line plot shows change in mean serum ALT level with time. Bar graph depicts distribution of ALT elevation according to severity at each time interval studied. Note: Normal ALT = 3–30 U/L; mild elevation in ALT = less than three times normal; moderate elevation in ALT = three to 20 times normal; severe elevation in ALT = greater than 20 times normal. Hash marks on x-axis indicate switch from days to weeks.

 
Mean bilirubin value at baseline was 0.6 mg/dL ± 0.53 (10.26 µmol/L ± 9.06). Bilirubin values increased in 44 (92%) of 48 procedures. Mean increase in bilirubin was 0.67 mg/dL (11.46 µmol/L) relative to baseline, and bilirubin returned to normal for all but five (13%) of 38 procedures at 1–2 weeks (Fig 5). Bilirubin levels remained less than 1.2 mg/dL (20.52 µmol/L) after 41 (85%) of 48 procedures. Overall, there were minor changes in bilirubin values that were, however, statistically significant (P < .001; median change, +0.10 mg/dL [1.71 µmol/L]).


Figure 5A
View larger version (11K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a: Graph shows (a) mean serum bilirubin, (b) mean serum ALKP, (c) mean serum myoglobin, and (d) mean hematocrit at different time intervals. Bars represent 95% confidence intervals. To convert serum bilirubin values to Système International units (micromoles per liter), multiply by 17.1. Hash marks on x-axis indicate switch from days to weeks.

 

Figure 5B
View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b: Graph shows (a) mean serum bilirubin, (b) mean serum ALKP, (c) mean serum myoglobin, and (d) mean hematocrit at different time intervals. Bars represent 95% confidence intervals. To convert serum bilirubin values to Système International units (micromoles per liter), multiply by 17.1. Hash marks on x-axis indicate switch from days to weeks.

 

Figure 5C
View larger version (11K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5c: Graph shows (a) mean serum bilirubin, (b) mean serum ALKP, (c) mean serum myoglobin, and (d) mean hematocrit at different time intervals. Bars represent 95% confidence intervals. To convert serum bilirubin values to Système International units (micromoles per liter), multiply by 17.1. Hash marks on x-axis indicate switch from days to weeks.

 

Figure 5D
View larger version (11K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5d: Graph shows (a) mean serum bilirubin, (b) mean serum ALKP, (c) mean serum myoglobin, and (d) mean hematocrit at different time intervals. Bars represent 95% confidence intervals. To convert serum bilirubin values to Système International units (micromoles per liter), multiply by 17.1. Hash marks on x-axis indicate switch from days to weeks.

 
Jaundice and/or pain were seen in three patients. One patient developed painless jaundice associated with a peak bilirubin level of 6.1 mg/dL (104.31 µmol/L) at 6–12 hours that normalized at 7 days. Another patient developed painless jaundice with a peak bilirubin level of 8.8 mg/dL (150.48 µmol/L) at 7 days that returned to normal 15 days after the procedure. No cause for hyperbilirubinemia could be identified in these two patients. The third patient developed right upper quadrant pain and a peak bilirubin level of 6.2 mg/dL (106.02 µmol/L) (baseline, 3.8 mg/dL [64.98 µmol/L]) 7 days after the procedure. An unenhanced computed tomographic (CT) scan showed high-attenuation material in the gallbladder and biliary duct dilatation; hemobilia was postulated as the cause of the elevated bilirubin value. This patient required an additional day in hospital, transfusion of 3 units of packed red blood cells, and readmission 5 days after the initial procedure because of persistent jaundice. Jaundice eventually resolved with observation alone.

Mean ALKP value at baseline was 99.58 U/L ± 97.29. ALKP values increased above baseline for 32 (67%) of 48 procedures; for 26 of these procedures, ALKP values peaked at 1 week. After the remaining 16 procedures, ALKP values decreased. There was significant change in ALKP levels at 0–6 hours (P < .001; median change, –5.0 U/L) and at 1–2 weeks (P < .001; median change, +29.0 U/L) (Fig 5).

Myoglobin levels were significantly higher than baseline values (P = .004; median change, +24.4 µg/L). Myoglobin levels increased above normal (0–100 µg/L) after 21 (44%) of 48 procedures. The mean baseline value was 41.7 µg/L ± 30.1; the peak was seen at the 0–6-hour interval, with a mean peak of 291 µg/L (range, 13.7–5180 µg/L; mean excluding one large outlier value, 183.4 µg/L), and myoglobin level returned to baseline at 1–2 weeks with no significant changes (P = .139) (Fig 5). Only one procedure resulted in severe myoglobinemia (5180 µg/L at 0–6 hours). This patient was treated prophylactically with intravenous infusion of d-mannitol and sodium bicarbonate for 24 hours. Serum myoglobin decreased to 860 µg/L, and there was no associated increase in serum creatinine level.

Mean creatinine value at baseline was 0.9 mg/dL ± 0.3 (79.56 µmol/L ± 26.52). There were no significant elevations in serum creatinine values at 0–6 hours (P = .72) or at 1–2 weeks (P = .74).

Hematologic Laboratory Tests and Complications
Mean platelet count at baseline was (246.0 ± 73.3) x 109/L. There was a significant decrease in platelet count from baseline values (P < .001) at 0–6 hours and 6–12 hours, but, when adjusted for baseline platelet counts, there were no significant changes (P = .22 and P = .74, respectively). The nadir in platelet count was at 12–24 hours in 24 (50%) of 48 procedures, at 6–12 hours after 13 (27%) procedures, at 0–6 hours after six (12%) procedures, and at 1–2 weeks in five (10%) procedures (percentages do not add up to 100% because of rounding). The mean nadir for the entire group was seen at 12–24 hours (Fig 6). The degree of change in platelet count for the entire group, as measured by the mean percentage decrease from baseline, was 20.8%, 32.4%, and 34.7% at 0–6 hours, 6–12 hours, and 12–24 hours, respectively. Mean maximal percent decrease in platelet count ({Delta}max percent platelet count) from baseline was 38% (absolute value, 92.3 x 109/L). The mean platelet count for the entire group returned to baseline at 1 week, with no significant difference between baseline and 1-week values (P = .62).


Figure 6
View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6: Line plot shows mean platelet count with time. Bar graph depicts distribution of platelet count at each time interval studied. Hash marks on x-axis indicate switch from days to weeks.

 
Platelet counts decreased below 100 x 109/L after three procedures. All three procedures resulted in large ablative margin volumes (74.7, 90.7, and 102.3 cm3; mean, 89.2 cm3 ± 13.9) that were significantly different from the mean ablative margin volume of all other ablation procedures (46.1 cm3 ± 35.5) (P = .01). After one procedure, one patient's platelet count decreased to 81.0 x 109/L at 12–24 hours, which necessitated an extra day in the hospital. After two other procedures, the platelet count fell below 50 x 109/L. In one case, the patient remained asymptomatic but was observed for 1 additional day in the hospital. In the other, the value decreased below 20 x 109/L as a result of DIC. The platelet count was 15 x 109/L at 6–12 hours and reached a nadir of 10 x 109/L at 12–24 hours. DIC was diagnosed on the basis of elevated prothrombin time, partial thromboplastin time, and D-dimer levels and a concurrent decrease in both fibrinogen level and platelet count. There was perihepatic hemorrhage on a contrast material–enhanced CT study performed 2 days after the procedure, with no active extravasation of contrast material. This patient continued to have low hematocrit and platelet count and required multiple transfusions of packed red blood cells and platelets. A contrast-enhanced CT study was repeated 6 days after the procedure and showed active contrast material extravasation within the ablation zone. Subsequent hepatic arteriography showed a pseudoaneurysm within the cryoablation zone that was embolized. During an 8-day hospitalization, the patient recovered completely but also was treated with 15 units of packed red blood cells, 14 units of platelets, 14 units of fresh-frozen plasma, and 2 units of cryoprecipitate.

WBC count did not increase significantly from baseline values ([6.03 ± 1.97] x 109/L) (P = .21). WBC count peaked at 6–12 hours ([9.17 ± 3.18] x 109/L) and increased above 11 x 109/L for only 13 (27%) of 48 procedures. WBC count returned to baseline at 1–2 weeks.

Hematocrit levels decreased significantly after all but one procedure (P = .001; median change, –2.9). Mean hematocrit value at baseline was 38.4% ± 4.3; the largest mean change in hematocrit (5%) was at 6–12 hours (Fig 5) and occurred in the absence of hemorrhage. There was a significant difference between baseline and 1–2 week values (P = .001; median change, –2.1). After two procedures, an additional day of hospitalization was needed for low hematocrit values (26% and 27%); 1 unit of packed red blood cells was administered after the first procedure.

Length of Hospital Stay
The average hospital stay for 48 procedures was 1.3 days. Patients were discharged on postprocedural day 1 after 41 procedures, day 2 after six procedures, and day 8 after one procedure.

Correlation of Ablation Volumes and Maximal Changes in Laboratory Values
Ablation zone diameters ranged from 3.4 to 8.4 cm, with a mean size of 6 cm. Mean ablation zone volume was 67.3 cm3 ± 41.2 (range, 7.3–191.4 cm3), and mean ablative margin volume was 48.8 cm3 ± 36.0 (range, 0–121.5 cm3). According to the generalized linear model, the ablative margin volume was predictive of changes in 0–6-hour values of AST (P = .02), ALT (P = .003), and myoglobin (P < .001) but was not predictive (P > .05) of changes in 0–6-hour values of platelet count; ALKP, bilirubin, and creatinine levels; hematocrit; and WBC count. The ablative margin volume was also predictive of the 1–2-week value of ALKP (P = .015). Tumor volume was not predictive of any parameter except the 0–6-hour and 1–2-week levels of ALKP (P < .001 and P = .001, respectively). Percent maximal decrease in platelet count correlated significantly with peak changes in AST (r = 0.72) and ALT (r = 0.68) (P < .001 for both) (Fig 7). Of note, the ablative margin volume in the one case with severe myoglobinemia was 121.5 cm3, the highest observed. The ablation zone did not include the diaphragm or intercostal muscles.


Figure 7A
View larger version (9K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7a: Scatterplot shows (a) correlation between peak rise in serum AST and (b) peak rise in serum ALT with percent maximal decrease in platelet count. Bold line = regression line, dotted lines = 95% confidence intervals.

 

Figure 7B
View larger version (9K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7b: Scatterplot shows (a) correlation between peak rise in serum AST and (b) peak rise in serum ALT with percent maximal decrease in platelet count. Bold line = regression line, dotted lines = 95% confidence intervals.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 IMPLICATIONS FOR PATIENT CARE
 References
 
Abnormalities in biochemical and hematologic parameters are common following percutaneous cryoablation of liver tumors; these parameters are most abnormal during the 0–6-hour period, and such abnormalities almost always resolve spontaneously. These findings are similar to those for cryoablation performed during open surgery. However, absolute AST values are typically higher after cryosurgery, during which larger probes result in larger volumes of tissue destruction (3,5–8,10). In addition, reports of cryosurgery have included cryoablation of larger tumors in combination with hepatic resection and vascular occlusion, as well as extrahepatic procedures (eg, cholecystectomy), each of which may contribute to elevated AST and ALT levels (3,7–10,15).

Our findings confirm what was suspected after cryosurgery; the magnitude of AST and ALT elevation relates to the amount of nontumoral liver parenchyma ablated (ie, the volume of normal liver parenchyma in the ablative margin volume) (8,10). Elevation in AST and ALT levels reflects this hepatocellular damage. Inclusion of normal liver in the ablation volume is needed to maximize the chance of treating the entire tumor; this is akin to a surgical margin during resection of liver tumors.

Our analysis showed that although small elevations in bilirubin levels were common, severe hyperbilirubinemia was uncommon. Mild self-limited bilirubinemia probably occurs as a result of ablation of small biliary ducts, whereas marked or persistent hyperbilirubinemia signifies larger duct injury. Our experience is similar to that in a study of 26 patients who underwent open cryosurgery (3); in that study, bilirubin levels higher than 3.0 mg/dL (51.3 µmol/L) were seen in only three patients, who had no clinical consequences. However, markedly elevated bilirubin levels could indicate a serious complication in the early postprocedure period; we observed one case of hemobilia. In the late postprocedure period, biliary stricture should be considered. Tumor volumes were predictive of ALKP levels; a satisfactory explanation could not be found for this observation. Ablative margin volumes were predictive of the increase in ALKP at 1–2 weeks. The delayed increase in ALKP and the relationship to the ablative margin volume seen in our study is consistent with the pattern expected for mechanical or functional cholestasis resulting from acute hepatocellular and bile duct injury (16). However, we found no clinical impact of persistently elevated ALKP levels. An earlier study (7) found no elevations in ALKP levels following cryosurgery, probably because the values were measured too soon to detect the delayed rise.

The mechanism for myoglobinemia after ablation is unclear (17). Liver tissue does not contain substantial amounts of myoglobin. Ablation of muscle tissue adjacent to the targeted lesion is a possible mechanism, but myoglobinemia (including the most severe case) was observed after procedures in which the ablation zone did not include muscle. Alternatively, cytokines released from ablated liver tissue may injure skeletal muscle, but this remains unproven (4). Early in our experience, we instituted prophylaxis for all patients undergoing percutaneous cryoablation to prevent previously reported myoglobinemia and myoglobinuria (4,5,9). When we observed that severe myoglobinemia (>1000 µg/L) was very uncommon, we changed our protocol and administered the prophylactic regimen only to patients with myoglobin levels greater than 1000 µg/L. We did not evaluate the clinical effectiveness of prophylactically treating patients with severe myoglobinemia. To our knowledge, there are no specific data on the prevention of myoglobinemia-induced renal failure after cryoablation. Our regimen was based on the opinion of our nephrologists and on extrapolation from literature on the prevention of renal failure in rhabdomyolysis associated with trauma (18).

The nadir in platelet count in our study occurred earlier than after cryosurgery (within 24 hours vs 48–72 hours) (2,4). This difference may relate to different study designs, as our study included more time points during the first 24 hours, and we were able to record earlier changes. Although thrombocytopenia was common, bleeding related to thrombocytopenia and DIC was rare. In our study, coagulopathy (one [2%] of 48 procedures) was less common than in cryosurgery (11 [3.8%] of 291 patients) and less severe than in cryosurgery—three deaths have been reported as a direct consequence of the coagulopathy (1). The mechanism of thrombocytopenia following cryotherapy is not well understood, but it may relate to a systemic inflammatory response to cryoablation (15) or to platelet aggregation and sequestration in the cryoablation zone (19).

The increase in WBC count was not statistically significant and was not predictive of infection or other complications. Mild rise in WBC count may reflect a general inflammatory response to tissue necrosis. Recently, other inflammatory markers such as interleukin-2, interleukin-6, C-reactive protein, and tumor necrosis factor-{alpha} have been reported to be elevated after cryoablation (12,15). These markers were not measured in our patients, and, because our study was retrospective, we could not evaluate their relevance.

The decrease in hematocrit levels was likely due to intraprocedural blood loss and hemodilution from administered fluids. Our data show that hematocrit typically decreases after cryoablation procedures, and, therefore, a mild decrease in hematocrit cannot be used alone to indicate hemorrhage.

AST levels have been found to correlate with platelet counts after hepatic cryosurgery (2). In our study, the ablative margin volume was not predictive of changes in the absolute platelet count, but the percent decrease in platelet count correlated strongly with peak rise in AST and ALT levels. This suggests that AST and ALT levels may serve as markers of as yet unidentified pathogenic substances released from ablated liver tissue that result in thrombocytopenia. There are additional factors that volume measurement alone may not be able to capture (eg, vascularity of the ablated region and surrounding tissues that may influence the systemic dissemination of locally released cytokines) that are better reflected by rise in AST and ALT levels. Tumor necrosis factor-{alpha} has been shown to activate caspases in mice that may be responsible for platelet consumption (20). Tumor necrosis factor-{alpha} infusions for the treatment of melanoma have been associated with thrombocytopenia (21). However, further studies would be needed to elucidate if these cytokines are markers for tissue destruction, if they are intermediate steps in a cascade that leads to thrombocytopenia, or if they directly result in platelet consumption. Because platelet counts decreased approximately 18 hours after the peak rise in AST and ALT, early monitoring of AST and ALT may be useful to predict thrombocytopenia and prevent bleeding complications in patients with severe elevations of AST and ALT. As in other studies (2), our results suggest that AST levels during the first 6 hours after the procedure are a good predictor of a subsequent decline in platelet count. This is especially important given our shorter length of stay (1.3 days), compared with a mean of 6.8 days in one surgical series (3) and a median stay of 10 days in another (9). Identifying a severely elevated AST level within 6 hours of the procedure should prompt consideration of a longer hospital stay, admission if the procedure is performed as an outpatient procedure, or closer follow-up in the 2–3 days after discharge.

The limitations of our study included a small sample size, although, to our knowledge, ours is the largest series to date on percutaneous cryoablation of liver tumors. Hemorrhage, secondary to thrombocytopenia or DIC, was less common in our study than in surgical series (1), but this finding needs to be interpreted with caution as our patient population was not large enough to allow definitive conclusions about such infrequent events.

Acute self-limited changes in biochemical and hematologic tests occur after percutaneous cryoablation of liver tumors; the severity of these changes correlate with the ablative margin volume and ablation zone volume. On the basis of our results, many laboratory changes, such as mild decreases in hematocrit and mild elevations in WBC count, are expected results of cryoablation. However, several laboratory findings also may be important indicators of a complication of the procedure. In particular, liver function tests, platelet count, and serum myoglobin levels are important tests in the care of patients after percutaneous cryoablation. On the basis of our study results, we continue to measure AST values within the first 6 hours following percutaneous cryoablation of liver tumors. If these levels are markedly elevated (>600 U/L), we observe the patient closely for subsequent thrombocytopenia (Table 2). We also closely follow patients with severe bilirubinemia (>3.0 mg/dL [51.3 µmol/L]) until resolution. When severe myoglobinemia (>1000 µg/L) is encountered in the first 6 hours after percutaneous cryoablation, we initiate prophylaxis (intravenous d-mannitol [0.3 mg/kg] and sodium bicarbonate [three ampules of 50 mEq each in 1 L of 5% dextrose in water at 150 mL per hour for 24 hours]) to prevent myoglobinuria-induced renal failure.


View this table:
[in this window]
[in a new window]

  		Table 2. Recommendations for Laboratory Test Utilization after Percutaneous Cryoablation of Liver Tumors

 

   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
IMPLICATIONS FOR PATIENT CARE
 References
 


   IMPLICATIONS FOR PATIENT CARE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 IMPLICATIONS FOR PATIENT CARE
References
 


   FOOTNOTES
 

Abbreviations: ALKP = alkaline phosphatase • ALT = alanine aminotransferase • AST = aspartate aminotransferase • DIC = disseminated intravascular coagulation • FAME = Fast Acquisition with Multiphase Efgre3D • WBC = white blood cell

2 Current address: Department of Radiology, St Joseph's Hospital and Medical Center and Arizona State University, Phoenix, Ariz Back

3 Current address: Department of Radiology, University of Massachusetts Medical Center, Worcester, Mass Back

See Materials and Methods for pertinent disclosures.

Author contributions: Guarantor of integrity of entire study, R.T.N.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, R.T.N., S.G.S., S.S.; clinical studies, R.T.N., S.G.S., K.T., E.v.S., S.S.; statistical analysis, R.T.N., N.A.O.; and manuscript editing, all authors


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 IMPLICATIONS FOR PATIENT CARE
 References
 

  1. Seifert JK, Junginger T, Morris DL. A collective review of the world literature on hepatic cryotherapy. J R Coll Surg Edinb 1998;43:141–154.[Medline]
  2. Cozzi PJ, Stewart GJ, Morris DL. Thrombocytopenia after hepatic cryotherapy for colorectal metastases: correlates with hepatocellular injury. World J Surg 1994;18:774–776.[CrossRef][Medline]
  3. Goodie DB, Horton MD, Morris RW, Nagy LS, Morris DL. Anaesthetic experience with cryotherapy for treatment of hepatic malignancy. Anaesth Intensive Care 1992;20:491–496.[Medline]
  4. Hamad GG, Neifeld JP. Biochemical, hematologic, and immunologic alterations following hepatic cryotherapy. Semin Surg Oncol 1998;14:122–128.[CrossRef][Medline]
  5. Onik G, Rubinsky B, Zemel R, et al. Ultrasound-guided hepatic cryosurgery in the treatment of metastatic colon carcinoma: preliminary results. Cancer 1991;67:901–907.[CrossRef][Medline]
  6. Ravikumar TS, Kane R, Cady B, Jenkins R, Clouse M, Steele G Jr. A 5-year study of cryosurgery in the treatment of liver tumors. Arch Surg 1991;126:1520–1523.[Abstract/Free Full Text]
  7. Shafir M, Shapiro R, Sung M, Warner R, Sicular A, Klipfel A. Cryoablation of unresectable malignant liver tumors. Am J Surg 1996;171:27–31.[CrossRef][Medline]
  8. Stewart GJ, Preketes A, Horton M, Ross WB, Morris DL. Hepatic cryotherapy: double-freeze cycles achieve greater hepatocellular injury in man. Cryobiology 1995;32:215–219.[CrossRef][Medline]
  9. Weaver ML, Atkinson D, Zemel R. Hepatic cryosurgery in treating colorectal metastases. Cancer 1995;76:210–214.[CrossRef][Medline]
  10. Zhou XD, Tang ZY, Yu YQ, Ma ZC. Clinical evaluation of cryosurgery in the treatment of primary liver cancer: report of 60 cases. Cancer 1988;61:1889–1892.[CrossRef][Medline]
  11. Dohi M, Harada J, Mogami T, Fukuda K, Toyama Y, Kashiwagi H. MR-guided percutaneous cryotherapy of malignant liver tumor under horizontal-magnetic open system: initial experience. J Hepatobiliary Pancreat Surg 2003;10:360–365.[CrossRef][Medline]
  12. Huang A, McCall JM, Weston MD, et al. Phase I study of percutaneous cryotherapy for colorectal liver metastasis. Br J Surg 2002;89:303–310.[CrossRef][Medline]
  13. Silverman SG, Tuncali K, Adams DF, et al. MR imaging-guided percutaneous cryotherapy of liver tumors: initial experience. Radiology 2000;217:657–664.[Abstract/Free Full Text]
  14. Mala T, Edwin B, Samset E, et al. Magnetic-resonance-guided percutaneous cryoablation of hepatic tumours. Eur J Surg 2001;167:610–617.[CrossRef][Medline]
  15. Seifert JK, Stewart GJ, Hewitt PM, Bolton EJ, Junginger T, Morris DL. Interleukin-6 and tumor necrosis factor-alpha levels following hepatic cryotherapy: association with volume and duration of freezing. World J Surg 1999;23:1019–1026.[CrossRef][Medline]
  16. Dufour RD. Evaluation of liver function and injury. In: Henry JB, ed. Clinical diagnosis and management by laboratory methods. 20th ed. Philadelphia, Pa: Saunders, 2001; 264–280.
  17. Shankar S, Tuncali K, vanSonnenberg E, Seifter JL, Silverman SG. Myoglobinemia after CT-guided radiofrequency ablation of a hepatic metastasis. AJR Am J Roentgenol 2002;178:359–361.[Free Full Text]
  18. Vanholder R, Sever MS, Erek E, Lameire N. Rhabdomyolysis. J Am Soc Nephrol 2000;11:1553–1561.[Free Full Text]
  19. Pistorius GA, Alexander C, Krisch CM, Feifel G, Schilling MK, Menger MD. Local platelet trapping as the cause of thrombocytopenia after hepatic cryotherapy. World J Surg 2005;29:657–660.[CrossRef][Medline]
  20. Piguet PF, Vesin C, Da Kan C. Activation of platelet caspases by TNF and its consequences for kinetics. Cytokine 2002;18:222–230.[CrossRef][Medline]
  21. Renard N, Nooijen PT, Schalkwijk L, et al. VWF release and platelet aggregation in human melanoma after perfusion with TNF alpha. J Pathol 1995;176:279–287.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nair, R. T.
Right arrow Articles by Shankar, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nair, R. T.
Right arrow Articles by Shankar, S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE