Sampling lymphoid tissue cells by ultrasound-guided fine needle aspiration of lymph nodes in HIV-infected patients

Bart, Pierre-Alexandreb; Meuwly, Jean-Yvesc; Corpataux, Jean-Marcd; Yerly, Sabinee; Rizzardi, Paolob; Fleury, Sylvainb; Munoz, Miguela; Knabenhans, Christianb; Welbon, Craigb; Pantaleo, Giuseppeb; Meylan, Pascal R.A.a,b; the Swiss HIV Cohort Study

Clinical: Original Papers

Objective: To establish the feasibility of using ultrasound-guided lymph node needle aspiration as a means to obtain lymphoid tissue cells for the determination of a series of immunologic and virologic measures in HIV-infected patients.

Design: First, a comparison of the characteristics of cell populations obtained by simultaneous needle aspiration and standard excisional biopsy in six patients. Second, use of lymph node needle aspiration to assess longitudinally T-cell subset changes in patients initiating highly effective antiretroviral treatment.

Methods: T-cell subsets (CD4 and CD8) and percentage Ki67+ cycling T cells were measured in lymph node cell populations harvested by ultrasound-guided aspiration or standard biopsy by flow cytometry. Cellular RNA content was assessed by a modification of the Roche Amplicor HIV-1 Monitor test.

Results: CD4 and CD8 T-cell percentage and HIV RNA cell content of lymph node cell suspensions obtained from the simultaneous performance of ultrasound-guided needle aspiration and excisional biopsy in the same patients were correlated (n=6). Among the 87 aspiration sessions reported here, mononuclear cell suspensions were obtained in 100% of the sessions, in numbers ranging between 4×104 to 6.7×106 cells (median: 7×105). This limited number of cells did not allow to perform all type of analyses in all patients. By prioritizing the cells for the determination of T-cell subsets and proliferation rate, this approach was instrumental for demonstrating the normalization of the T-cell subset ratio and the kinetic of normalization of proliferating rates of CD4 and CD8 T cells, as well as the decrease in HIV-1 viral load in the lymph node following HAART initiation.

Conclusion: Ultrasound-guided aspiration appears to be a non-invasive and ad libitum, safe and repeatable procedure for the longitudinal monitoring of changes in lymph nodes.

Author Information

From the aInstitute of Microbiology, bDivision of Infectious Diseases, cDepartment of Radiology and dDepartment of Surgery, Centre Hospitalier Universitaire Vaudois, University of Lausanne and eLaboratoire de Virologie, Division des Maladies Infectieuses, Hôpital Cantonal Universitaire, Genève, Switzerland. *See Appendix.

Sponsorship: This study has been financed by the Swiss Federal Office of Public Health (Grant no 3600.010.1 # 184 to the Swiss HIV Cohort Study) and the FNS Grant no 3239-47330.96.

Note: Part of this work was presented at the 10th World AIDS Conference, Geneva, Switzerland, June 29-July 4 1998, [Abstract 41254].

Correspondence to Pascal R.A. Meylan, Institut de Microbiologie, Centre Hospitalier Universitaire Vaudois, CH-1011 Lausanne, Switzerland.

Received: 19 January 1999; revised: 17 May 1999; accepted: 25 May 1999.

Article Outline
Back to Top | Article Outline


In recent years, the central role of the lymphoid tissue in HIV pathogenesis has been recognized both as the compartment where the vast majority of lymphoid cells reside [1] and as the place where the bulk of virus production (productive infection of CD4 T lymphocytes) and storage (viral particles trapped on follicular dendritic cells) occur [2-4]. Changes of virus load in the lymphoid tissue in response to antiretroviral treatment, though somewhat slower, have been shown to correlate with changes in the blood [5-9]. In contrast, discrepancies have been noted with respect to the virus load in the lymphoid tissue versus blood compartment, in particular in asymptomatic untreated patients [2]. Discrepancies have also been noticed regarding T-cell populations, in particular with higher proportions of CD4 T cells in the lymphoid compartment compared with the blood of HIV-infected patients [10,11]. Differences in percentage of cycling (Ki67+) T cells have also been noted between lymphoid tissue and blood cell populations [10]. This makes sampling of the lymphoid tissue central to the understanding of the HIV pathogenesis and the response to treatment.

The standard procedure to sample lymphoid tissue is excisional biopsy of a lymph node. This procedure, however, cannot be repeated ad libitum for obvious practical and ethical reasons. These considerations led us to test needle aspiration of lymph nodes as a less invasive procedure to sample cells from the lymphoid tissue. Using finger-guided needle aspiration, we have previously established that cells can be obtained in about 50% of the patients in numbers sufficient to perform immunophenotyping of the mononuclear cell suspensions by flow cytometry, to measure the RNA and DNA cell content by quantitative polymerase chain reaction (PCR), and the cell-associated infectivity by semiquantitative cultures [11]. In a follow-up study, we demonstrated the possibility with this approach to monitor longitudinally changes in the lymphoid tissue induced by antiviral treatment [8].

In the present study, we show the increased success rate of lymph node needle aspiration in obtaining cells from the lymphoid tissue when performed under ultrasonic guidance.

Back to Top | Article Outline



HIV-infected patients enrolled in two prospective studies testing the effect of HAART on various aspects of the immune system were eligible for needle aspiration studies if they gave written informed consent to undergo needle aspiration in accordance with the Hospital Ethical Committee guidelines. The first study (AVIB) is a phase II randomized study of changes in T-cell turnover, immune function and lymph node architecture during long-term suppression of viraemia associated with combination antiviral therapy [abacavir (1592U89) plus nelfinavir plus saquinavir versus abacavir plus amprenavir (141W94) plus nelfinavir] alone or antiviral plus immune-based (subcutaneous interleukin-2 or Remune, an inactivated virus preparation in adjuvant; Immune Response Corporation, Carlsbad, CA, USA) therapy in antiretroviral naive HIV-1 infected subjects with a CD4 T-cell count >250×106cells/l. In this study, at base-line, patients underwent concomitant excisional lymph node biopsy and needle aspiration that allowed comparison of the data obtained with the two procedures. On weeks 18, 33 and 53 after HAART initiation, patients underwent repeat aspirations. The second study (CNAB 2006) is a phase II study testing the safety and the efficacy of abacavir (1592U89) plus amprenavir (141W94) on viral load and on T-cell changes in blood and in lymph nodes in naive patients with CD4 cell count >400×106cells/l and plasma viremia>5000HIV RNA copies/ml. In the latter study, patients underwent excisional lymph node biopsy at base-line and ultrasound-guided needle aspirations at weeks 24, 36 and 48 after treatment initiation to provide longitudinal analyses regarding the changes in the lymphoid tissue following HAART.

Back to Top | Article Outline
Ultrasound-guided lymph node aspiration

Needle aspiration was performed by inserting a 22-G needle in the cortical region of a right inguinal or femoral lymph node under guidance of an Aloka SSD 2000 machine (Aloka, Tokyo, Japan) with a linear probe at 7.5MHz. Doppler analysis was used to localize and avoid blood vessels, including lymph node medullary vessels. Vacuum was applied with a syringe holder for about 1 minute with a 20ml syringe connected to the needle through a 40cm×2mm arteriography connection tube (Connectub Artério moyenne pression, polyethylene, Plastimed, 95321 St-Leu-la-Forêt, France). After aspiration, the vacuum was broken before the needle was withdrawn from the lymph node and the needle rinsed in 500μl of ACD AB-16 anticoagulant solution diluted 1/10 in PBS. Aspiration was usually performed twice on the same node and the cells obtained pooled for analysis. An aliquot was stained with Türck solution (1% v/v acetic acid, 0.01% w/v crystal violet in distilled water) and the leukocytes enumerated microscopically.

Back to Top | Article Outline
Cell sample analysis procedures

The cells obtained were subjected to a hypotonic shock to eliminate red blood cells, washed, and depending on the aspiration yield, the cells were appropriately aliquoted in order to perform T-cell subset phenotype (CD3, CD4 and CD8) and Ki67+ expression as described [10]. While 50000 cells were sufficient to perform the T-cell subset characterization, about 500000 cells were required for the intracellular Ki67 antigen. When sufficient number of cells were available, they were pelleted and stored at -80°C until the determination of HIV gag RNA using the Roche Amplicor HIV-1 Monitor test (Roche, Basel, Switzerland). For the determination of the HIV-1 gag RNA, cell samples (5×105cells) were lysed in the lysis buffer containing the RNA internal quantitation standard (IQS). The nucleic acids were then extracted, treated with RNase-free DNase I and an amount of extract corresponding to 125000cells was amplified in duplicate according to the manufacturer‚s instructions with modifications as recently described [12]. Excisional lymph node biopsies and isolation of mononuclear cells were performed as described [10].

Back to Top | Article Outline
Statistical analysis

Statistical analysis was performed with the tests indicated in the result section, using the Graphpad Prism software (San Diego, California, USA).

Back to Top | Article Outline


Comparison between excisional biopsy and ultrasound-guided needle aspiration samples

Determination of CD4 and CD8 T-cell percentages were performed in mononuclear cell suspensions obtained simultaneously from excisional lymph node biopsy and from ultrasound-guided lymph node aspiration in the controlateral node of the same patient. Figure 1 demonstrates the location of the aspiration needle within the cortical region of an inguinal node, i.e. the location where aspiration was performed. Figure 2 shows the comparative analysis performed in six subjects at baseline in the frame of the AVIB study (one patient had two aspiration samples processed separately and compared with the data of the same biopsy sample). A significant linear correlation was observed in both the percentage of CD4 and CD8 T cells isolated using these two procedures found in each type of cell population. CD4 T-cell percentage was similar in mononuclear cell suspensions from ultrasound-guided lymph node aspirates and excisional lymph node biopsy (see the correlation line slope=1, Fig. 2, upper panel). The percentage of CD8 T cells was slightly lower in aspiration compared with the biopsy samples (see the correlation line slope <1, Fig. 2, lower panel).

In five out of six subjects, sufficient numbers of cells were obtained for the determination of HIV cell- associated RNA by PCR analysis. As shown by Table 1, similar amounts of HIV RNA (105 to 2×105copies/106lymph node cells) were found in mononuclear cell suspensions. The differences observed by using the two procedures were within the range of variability of the PCR assay.

Back to Top | Article Outline
Longitudinal assessment of changes induced in lymphoid tissue by HAART

Fine needle aspiration of lymph nodes was used to assess the effect of HAART (abacavir (1592U89) and amprenavir (141W94)) on T-cell changes in the lymphoid tissue in naive patients (CNAB 2006) with CD4 cell count>400×106cells/l and plasma viremia>5000 HIV RNA copies/ml. After a lymph node biopsy performed at baseline, needle aspirations were planned at weeks 24, 36 and 48 of follow-up. Of 40 patients enrolled in this prospective study, 36 agreed to undergo aspiration. Two patients dropped out of the study before week 24, i.e. time of the first planned aspiration. The number of aspirations performed diminished at the later time points due to the fact that the present study was terminated before all patients had completed follow-up. Figure 3 shows the number of ultrasound-guided lymph node aspirations performed at the different time points of this study as of 31 July 1998, as well as the median and range of the number of cells obtained, and the number of specimens on which the various determinations could be performed. The cell number obtained in an aspiration session ranged from 4×104 to 6.7×106 (median: 7×105). About 50-100000 cells were used for the measure of the CD4 and CD8 T-cell percentage, and this determination could be performed in 100% (80 of 80) sessions. Figure 4a shows the evolution of T-cell subsets in 10 patients included in the AVIB study. The percentage of CD3 T cells in the lymphoid tissue increased slightly over the first year of treatment (P=0.30, one-way ANOVA), but actually reflected opposite changes of percentage CD4 and CD8 T cells. The percentage of CD4 T cells increased significantly over time (P=0.0011, one-way ANOVA), while CD8 T cells decreased, although not significantly (P=0.13, one-way ANOVA). Similar findings were observed in the frame of the CNAB 2006 study and are reported elsewhere [13].

For the determination of the percentage Ki67+ among T cells, 500000 cells were required and this determination could be performed in 69 of 80 (86%) of the sessions. Indeed, this assay was given priority over the HIV RNA assay. This approach was instrumental in demonstrating changes in proliferating (Ki67+) T-cell populations over the year following HAART initiation, compared with base-line levels. Proliferating populations were calculated from the percentage of Ki67+ T cells in cell suspensions obtained by lymph node aspiration as recently described [10]. As shown on Fig. 4b, the present approach allowed the percentage proliferation among T cells at various times after initiating HAART to be determined and comparison of these values with values obtained at base line by excisional biopsy. The total number of proliferating T cells (CD3+) in the lymphoid tissue remained stable over the first year of treatment (P=0.90, one-way ANOVA), but actually reflected opposite changes of CD4 and CD8 proliferating cells. Numbers of proliferating CD4 T cells increased over time, a difference that almost reached statistical significance (P=0.059, one-way ANOVA). This was compensated for by a significant decrease in the numbers of proliferating CD8 T cells (P=0.006, one-way ANOVA).

For the determination of the HIV RNA copy number in the cell suspension, 500000 cells were required, and enough cells were available for this assay in 39 of 80 (49%) sessions. Examples of changes of HIV-1 RNA load in lymph node cells of four patients are shown in Fig. 5: three out of four patients had undetectable HIV RNA (<3copies/106lymph node cells) from 24 to 48 weeks after treatment initiation.

Back to Top | Article Outline

Among the 87 aspiration sessions mentioned in this report, no adverse event (haemorrhage, infection) was noticed.

Back to Top | Article Outline


The lymphoid tissue is the central anatomic compartment for HIV replication and pathogenesis. Therefore, it is important to gain access to lymphoid tissue specimens in order to improve the understanding of HIV pathogenesis and the efficacy of antiretroviral therapy. Although surgical lymph node biopsy is the conventional approach to sample lymphoid tissue, providing the largest amount of biological material, its invasive nature and cost limits its wide use. Thus, the development of alternative, less invasive, sampling procedures is very important to allow repeated sampling of lymphoid tissue for the longitudinal assessment of virus load and of the cell populations changes in response to therapy.

On the basis of the present results, ultrasound-guided aspiration appears to be a non-invasive, safe, and ad libitum repeatable procedure. We show that this technology provides lymphoid tissue cell suspensions in a vast majority of non-selected patients, and thus represents a striking improvement over finger-guided aspiration [8,11]. In these respects, it compares favourably with surgical biopsy, core needle biopsy which provided samples in about 50% of the patients in a recent study [14], or tonsil biopsy [4,5].

The lymph node sampling procedure we describe here was instrumental for the immunophenotype analysis of the mononuclear cells and allowed the quantification of cell-associated HIV RNA. Of note, experimental evidence was provided that both immunologic and virologic measures obtained in cell suspensions isolated by this technology were similar to those in cell suspensions obtained from excisional lymph node biopsies.

While CD4 T-cell percentages in aspiration and excision samples were very similar, CD8 T-cell percentages, although significantly correlated, were somewhat lower in aspiration samples compared with excision samples (see Fig. 2). The reason for this discrepancy is unclear. Of note however, the aspiration was performed taking care to avoid the lymph node medullary blood vessels from fear of blood contamination. The aspirated samples thus represent a preferential sampling of the lymph node cortical and paracortical area avoiding the medullary region. CD8 T cells are greatly increased in HIV-infected lymph nodes [10]. It is conceivable that the proportion of CD8 T cells is not as increased at the site of the aspiration as compared with the whole lymph node when sampled by excisional biopsy. However, these slight differences do not seem to have a major impact on the analysis of the changes in CD4 T-cell percentage following HAART.

With regard to the determination of viral load, similar amounts of HIV RNA copies per 106 cells were measured in cell suspensions obtained by either approach. A two-fold difference was observed in patient 1003 with a higher viral load in the aspiration sample compared with the excisional sample. However, these differences are within the range of variability of the PCR assay. Compared with solid tissues biopsies, our procedure does not allow the extraction of RNA from whole tissue, but only from isolated cells. Lafeuillade et al. [15] have nicely demonstrated, extracting total tissue or cell suspensions obtained from excisional biopsies, that the pool of extracellular HIV-1 RNA is far greater (2-3 Logs) than that of lymph node mononuclear cells. Interestingly, the level of HIV-1 RNA found by these authors in lymph node mononuclear cells of drug-naive patients was similar (105-106copies/106 cells) to the level that we found in cell suspensions obtained by aspiration or excisional biopsies in the present study (see Table 1 and Fig. 5, time 0).

Limitations of ultrasound-guided lymph node aspiration include the technical skills required both for performing the aspiration and for the manipulation of samples containing small cell numbers, the lack of information on the lymph node architecture, and the variable, sometimes limited cell number yield. Indeed, the present approach generates a cell suspension, not a tissue fragment. Obviously, this precludes the assessment and evaluation of lymphoid tissue architecture that can be performed in lymph node, tonsil and rectal biopsies [2-5].

However, the analysis of the changes in viral load is certainly the most important information for evaluation of the effectiveness of antiretroviral therapy. The ultrasound-guided lymph node aspiration is a potentially very useful approach for these purposes. It is true that one or two aspirations may not provide the number of mononuclear cells necessary to perform both flow cytometry and determinations of viral load. It is important, however, to underscore that in the two clinical studies in which this procedure was applied [10] and [13], the priority was to assess the immunologic changes following HAART introduction. There are no technical or safety reasons to preclude the performance of multiple aspirations to obtain larger numbers of cells for both immunologic and virologic analyses.

In conclusion, evidence is provided that ultrasound-guided lymph node aspiration is an ad libitum repeatable procedure for the longitudinal monitoring of immunologic and virologic changes in lymph nodes following antiretroviral therapy.

Back to Top | Article Outline


1. Trepel F. Number and distribution of lymphocytes in man. A critical analysis. Klin Wschr 1974, 52: 511-515.
2. Pantaleo G, Graziosi C, Demarest JF, et al. HIV infection is active and progressive in lymphoid tissue during the clinically latent stage of disease. Nature 1993, 362: 355-358.
3. Embretson J, Zupancic M, Ribas JL, et al. Massive covert infection of helper T lymphocytes and macrophages by HIV during the incubation period of AIDS. Nature 1993, 362: 359-362.
4. Haase AT, Henry K, Zupancic M, et al. Quantitative image analysis of HIV-1 infection in lymphoid tissue. Science 1996, 274: 985-989.
5. Cavert W, Notermans DW, Staskus K, et al. Kinetics of response in lymphoid tissue to antiretroviral therapy in HIV-1 infection. Science 1997, 276: 960-964.
6. Cohen OJ, Pantaleo G, Holodniy M, et al. Decreased human immunodeficiency virus type 1 plasma viremia during antiretroviral therapy reflects downregulation of viral replication in lymphoid tissue. Proc Natl Acad Sci USA 1995, 92: 6017-6021.
7. Tamalet C, Lafeuillade A, Fantini J, Poggi C, Yahi N. Quantification of HIV-1 viral load in lymphoid and blood cells: assessment during four-drug combination therapy. AIDS 1997, 11: 895-901.
8. Burgisser P, Spertini F, Suter C, Pagani JL, Meylan PRA. Monitoring the response to antiviral treatment in the lymph nodes of HIV-infected patients by serial fine needle aspiration. J Infect Dis 1997, 175: 1202-1205.
9. Wong JK, Günthard HF, Havlir DV, et al. Reduction of HIV-1 in blood and lymph nodes following potent antiretroviral therapy and the virologic correlates of treatment failure. Proc Natl Acad Sci USA 1997, 94: 12574-12579.
10. Fleury S, De Boer RJ, Rizzardi GP, et al. Limited CD4+ T-cell renewal in early HIV-1 infection: effect of highly active antiretroviral therapy. Nature Med 1998, 4: 794-801.
11. Meylan PRA, Burgisser P, Weyrich-Suter C, Spertini F. Viral load and immunophenotype of cells obtained from lymph nodes by fine needle aspiration compared to peripheral blood cells in HIV-infected patients. J Acquir Immune Defic Syndr 1996, 13: 39-47.
12. Yerly S, Rutschmann OT, Opravil M, et al. Cell-associated HIV-1 RNA in blood as indicator of viral load in lymph nodes. J Infect Dis 1999, (in press).
13. Bart PA, Rizzardi GP, Tambussi G, et al. Quantitative normalization of CD4 cells in blood and lymph node of HIV-infected persons at early stage of chronic infection and no previous antiretroviral therapy treated with abacavir plus amprenavir. VI International Conference on Retroviruses and Opportunistic Infections. Chicago, January-February 1999. [abstract 626].
14. Harris M, Patenaude P, Cooperberg P, et al. Correlation of virus load in plasma and lymph node tissue in human immunodeficiency virus infection. J Infect Dis 1997, 176: 1388-1392.
15. Lafeuillade A, Poggi C, Tamalet C, Profizi N. Human immunodeficiency virus type 1 dynamics in different lymphoid tissue compartments. J Infect Dis 1997, 176: 804-806.
Back to Top | Article Outline

The members of the Swiss HIV Cohort Study are M.Battegay (Co-Chairman of the Scientific Board), E.Bernasconi, Ph.Bürgisser, M.Egger, P.Erb (Chairman of the Group ‚Laboratories‚), W.Fierz, M.Flepp (Chairman of the Group ‚Clinics‚), P.Francioli (President of the SHCS, Centre Hospitalier Universitaire Vaudois, CH-1011- Lausanne), H.J.Furrer, P.Grob, B.Hirschel (Chairman of the Scientific Board), L.Kaiser, B.Ledergerber, R.Malinverni, L.Matter, M.Opravil, F.Paccaud, G.Pantaleo, L.Perrin, W.Pichler, J.-C.Piffaretti, M.Rickenbach (Head of Data Center), P.Sudre, J.Schupbach, A.Telenti, P.Vernazza, R. Weber.


Ultrasound-guided; lymph node; needle aspiration; T-cell subsets; HAART; T-cell proliferation; non-invasive; lymph node; longitudinal study

© 1999 Lippincott Williams & Wilkins, Inc.