ORIGINAL RESEARCH
Rethinking Viscosupplementation
Ultrasound- Versus Landmark-Guided Injection for Knee
Osteoarthritis
Zachary T. Lundstrom, MD, Terin T. Sytsma, MD, Laura S. Greenlund, MD, PhD
Received March 7, 2019, from the Depart-
ment of Internal Medicine, Mayo Clinic,
Rochester, Minnesota, USA. Manuscript
accepted for publication June 9, 2019.
All of the authors of this article have
reported no disclosures.
Address correspondence to Laura S.
Greenlund, MD, PhD, Department of Internal
Medicine, Mayo Clinic, 200 First St SW,
Rochester, MN 55905 USA.
E-mail: greenlund.laura@mayo.edu
Abbreviations
BMI, body mass index; HA, hyaluronic
acid; OA, osteoarthritis; US, ultrasound
doi:10.1002/jum.15081
© 2019 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2019; 00:1–5 | 0278-4297 | www.aium.org
Objectives—Viscosupplementation, intra-articular injection of hyaluronic acid
(HA), for knee osteoarthritis has fallen somewhat out of favor, with studies fail-
ing to show consistent benefits in pain reduction. Hyaluronic acid must enter
the joint space to be beneficial; however, landmark-guided injection can be sub-
stantially inaccurate, especially in obese patients. We aimed to determine whether
ultrasound (US) guidance to ensure needle placement for HA knee injection
resulted in improved outcomes as reflected by surgery-free survival compared to
landmark-guided HA knee injection.
Methods—All community-dwelling patients in 6 contiguous surrounding counties
receiving HA knee injection either by landmark guidance (n = 647) or by US
guidance (n = 500) were analyzed for the degree of arthritis, body mass index,
follow-up injection, and subsequent need for knee arthroplasty. A subgroup analy-
sis of obese patients was also performed.
Results—The US- and landmark-guided HA injection cohorts were similar with
respect to sex, body mass index, and the degree of arthritis. Of 1147 patients
receiving knee HA injection, 462 subsequently underwent knee arthroplasty. Sig-
nificantly fewer patients in the US-guided HA injection cohort went to surgery
(33.2%) compared to the landmark-guided cohort (45.8%; P < .001). The sub-
group analysis for obese patients showed even larger differences (34.8% versus
51.8%; P < .001).
Conclusions—Knee osteoarthritis treatment by viscosupplementation can be opti-
mized by US guidance, ensuring intra-articular needle placement. Using an objec-
tive surgical outcome, our study shows that rethinking viscosupplementation to
ensure intra-articular delivery improves effectiveness. Patients receiving US-guided
knee HA injection were significantly less likely to undergo subsequent knee
arthroplasty than patients receiving landmark-guided HA injection.
Key Words—knee; osteoarthritis; treatment; ultrasound; viscosupplementation
T he prevalence of knee osteoarthritis (OA) is sharply rising
and at a younger age,1 but nonsurgical options adminis-
tered in the office are limited, especially those covered by
most insurance. Optimizing knee viscosupplementation by ensuring
intra-articular placement could help fill this gap. Past evidence of
benefits in pain reduction and improved function after viscosup-
plementation has been inconsistent.2–7 Most previous studies and
meta-analyses on viscosupplementation have used landmark knee
injections, which have varying accuracies, with reports as low as 43%
and an average of 79% being placed within the joint depending on
the approach used.8 For example, data from a large systematic
review showed landmark-guided accuracy of 70% and 74% when
Lundstrom et al—Ultrasound-Guided Knee Viscosupplementation
injecting anteriorly at the lateral and medial joint lines,
respectively, with the knee in a flexed position. Using
extended knee superolateral patellar and superomedial
patellar approaches yielded 87% and 82% accuracy,
respectively.8 In contrast, knee joint injection accuracy
is substantially improved (average 98% accurate in
1100 injections) with the help of image guidance.8,9
Hyaluronic acid (HA) injection for viscosupple-
mentation must be placed intra-articularly to be effec-
tive, and inadvertent extra-articular injection does not
provide a benefit. Ultrasound (US) visualization ensures
the accuracy of needle placement into the knee joint to
optimize viscosupplementation benefits, particularly
when landmarks are difficult to identify, as occurs in
obese patients. This study examined the effect of
ensuring intra-articular HA injection by using US visu-
alization compared to landmark-guided HA injection
on subsequent knee arthroplasty rates. The outcome
measure end point of knee arthroplasty is both objec-
tive and has substantial economic and morbidity impli-
cations. Several recent studies from large-claim data
bases in the United States and France showed that
HA injections are associated with delayed total knee
arthroplasty; however, no distinction was made between
landmark- and US-guided procedures, and no data on
patients receiving HA injection but not requiring sur-
gery were presented.10–14 A separate study using knee
arthroplasty as an outcome did not show a difference
between low- and high-molecular-weight HA in 30,417
propensity-matched patients.15
Materials and Methods
Study Participants
Patients residing in 6 contiguous surrounding counties
and receiving an HA knee injection (all formulations
were included) during the years 2008 to 2014, a time
frame that allowed an adequate postprocedure dura-
tion to determine the need for subsequent knee
arthroplasty, were eligible for inclusion. Given no
differences between low- and high-molecular-weight
HA on the arthroplasty rate in an earlier study,15 the
types of products were not distinguished in this study.
Patients referred by and treated in a variety of depart-
ments, including primary care, rheumatology, orthopedic
surgery, sports medicine, physical medicine and rehabili-
tation, and radiology, were initially identified by review
2 J Ultrasound Med 2019; 00:1–5
of the electronic medical record database. Individual
patient charts were examined retrospectively for demo-
graphic data, the treatment administered, radiographic
severity of OA, follow-up interventions, and the date of
surgery (unicompartmental or total knee arthroplasty) if
performed. In the US group 1.2% and in the landmark
group 1.9% of patients did not have follow-up visits.
Radiographic severity was graded by dedicated
musculoskeletal radiologists using an internal system
with uniformity across the radiology department. The
radiologists grading the radiographic severity were
blinded to (had no knowledge of) the method of HA
injection. The data extractors were also blinded to the
cohort when reviewing the reports. For landmark injec-
tion, the portal of entry used was not distinguished,
and for US injection, most procedures were performed
by a lateral suprapatellar bursa approach. The two
cohorts were followed for a median of 6.5 years (US-
guided, interquartile range, 5.2–7.2 years) and 6.9 years
(landmark-guided, interquartile range, 5.6–8.0 years)
after HA injection. All patients had given written con-
sent for participation, and the study was approved by
the Institutional Review Board.
Statistical Analysis
Patient demographic information was summarized by
descriptive statistics. Comparisons between cohorts
were by a 2-tailed t test for age and body mass index
(BMI) and by the Pearson χ2 test for categorical vari-
ables. P ≤ .05 was considered significant. The analysis
of surgery-free survival is presented in the form of
Kaplan-Meier curves and compared by the log-rank
test. Statistical analyses were performed with JMP
version 13 software (SAS Institute Inc, Cary, NC).
Results
Among 1147 individual community-dwelling patients
receiving an HA injection during a 6-year time frame
(2008-2014, timing that allowed an adequate post-
procedure duration to determine the need for subse-
quent knee arthroplasty), 647 of the procedures were
landmark guided, and 500 were US guided. The
median BMI, sex distribution, and radiographic sever-
ity of OA were comparable between the cohorts. The
mean age of the US-guided cohort was slightly less
than that of the landmark-guided cohort (Table 1).

Knee arthroplasty rates were significantly reduced
in the US-guided cohort (33.2%) compared to the
landmark-guided cohort (45.8%; P < .001, Pearson χ2
test). Differences in surgery-free survival between the
cohorts were seen early in the course of follow-up
(Figure 1). A subgroup analysis of obese patients (BMI
>30 kg/m2) showed more marked differences in the
arthroplasty rate, with 34.8% in the US-guided cohort
going to surgery versus 51.8% in the landmark-guided
cohort (P < .001, Pearson χ2 test), likely because of dif-
ficulty identifying landmarks in the obese patients,
resulting in less accurate landmark-guided HA injection.
Not only were knee arthroplasty rates less after
US-guided HA injection, but fewer patients in the
US-guided cohort (27.4%) received a subsequent cor-
ticosteroid injection than the landmark-guided cohort
(34.0%; P < .002), and they were more likely to
receive another HA injection (20.3% versus 15.0%;
Figure 1. Kaplan-Meier surgery-free survival curves with log-rank
analysis to compare cohorts (P ≤ .001).
Table 1. Characteristics of Community-Dwelling Patient Cohorts
Characteristic US Guided (n = 500)
Female 328 (65.6)
Age, y 60.6  14.8
BMI, kg/m2 31.9 (27.7–37.0)
BMI >30 kg/m2, % 60.6
Radiographically moderate/severe 361 (72.2)
Received repeated HA 102 (20.3)
Received CS during follow-up 137 (27.4)
Received surgery 166 (33.2)
Data are presented as number (percent), mean  SD, and median (interquartile range) where applicable. Comparisons were between
US-guided versus landmark-guided cohorts or surgery versus no-surgery cohorts by a 2-tailed t test for age and BMI and by the Pearson χ2
test for categorical variables. Unless designated, P values were nonsignificant (all P ≥ .2). CS indicates corticosteroid.
a
P ≤ .002.
J Ultrasound Med 2019
Lundstrom et al—Ultrasound-Guided Knee Viscosupplementation
P < .002), suggesting that their first HA injection
would be more effective if it were US guided
(Table 1). In patients receiving a subsequent HA
injection, most procedures were by the same method
(US or landmark guided), with only a small cross-
over: 4.2% for US- to subsequent landmark-guided
and 3.7% for landmark- to subsequent US-guided HA
injection.
Discussion
In our community-dwelling patients, ensuring intra-
articular placement of HA by US-guided injection
was associated with a significantly reduced knee
arthroplasty rate compared to landmark-guided HA
injection, particularly in obese patients. The mecha-
nisms of action for HA include biomechanical effects
of viscoelasticity and lubrication and physiologic
effects on synoviocytes and chondrocytes, providing
potentially sustained benefits.16–18 The difference in
knee arthroplasty rates between the cohorts in our
study was not explained by differences in age, BMI,
or radiographic severity. Prior large prospective
studies and meta-analyses using landmark HA injec-
tion have reported inconsistent benefits of vis-
cosupplementation, which could be explained in part
by inaccuracies of needle placement with inadvertent
injection outside the joint. In addition, most previ-
ous studies used subjective outcome measures such
as patient-reported pain scales. In contrast, our study
outcome measure was knee arthroplasty—an objec-
tive end point. Since the primary indication for knee
Landmark Guided (n = 647) Surgery (n = 462) No Surgery (n = 685)
426 (65.8) 300 (64.9) 454 (66.3)
63.4  13.3a 62.2  10.2 62.1  16.2
31.5 (27.5–36.1) 32.2 (28.6–36.8) 31.0 (27.0–36.3)
60.0 66.6 55.8a
478 (73.9) 364 (78.8) 476 (69.5)a
97 (15.0)a 95 (20.5) 149 (21.8)
220 (34.0)a 141 (30.3) 216 (31.6)
296 (45.8)a
3
Lundstrom et al—Ultrasound-Guided Knee Viscosupplementation
arthroplasty is pain and joint dysfunction, the rate of
surgery is likely an objective reflection of viscosupple-
mentation effectiveness. Although the decreased rate of
knee arthroplasty among patients receiving US-guided
viscosupplementation could theoretically be due to
a factor not accounted for, this seems unlikely.
Unaccounted-for biases in patient referral for US-
versus landmark-guided injection or biases in selec-
tion for knee arthroplasty are possible, but given that
referrals and HA injections in our study were by a
variety of sources, including primary care, rheuma-
tology, orthopedic surgery, sports medicine, physical
medicine and rehabilitation, and radiology, it seems
unlikely that any consistent bias affected referrals
among all departments. Certainly, factors such as
age, body habitus, severity of arthritis, and com-
orbidities contribute to these decisions about the
injection method and suitability for arthroplasty.
Our data are also consistent with studies of large-
claim databases that showed a delay in knee arthroplasty
associated with HA injection.10–14 Given that these
studies did not distinguish between the methods of HA
injection (US versus landmark guided), it is likely that
greater delays to surgery may have been observed in
the subset of patients receiving US-guided HA. In addi-
tion, these studies only reported information on the
subset of patients who went to surgery after knee HA
injection (no data were presented about patients who
did not need arthroplasty), whereas our study followed
and accounted for outcomes of all patients receiving
HA injections. To our knowledge, this was the first
large study comparing the knee arthroplasty rate after
US- versus landmark-guided injections of HA.
Options for in-office treatment of OA are cur-
rently limited, especially those covered by most insur-
ance, but the need continues to grow. Given the
expense of viscosupplementation products, ensuring
intra-articular delivery by US-guided injection is ideal.
Ultrasound guidance adds cost to the procedure;
however, given the obesity epidemic (making land-
marks more difficult to identify) and the rising preva-
lence of OA with onset at younger ages, rethinking
viscosupplementation and ensuring proper delivery
has potentially enormous benefits, both economic
and reduced morbidity.
In conclusion, our study shows that patients receiv-
ing US-guided knee HA injection were significantly less
4 J Ultrasound Med 2019; 00:1–5
likely to undergo subsequent knee arthroplasty than
patients receiving landmark-guided HA injection.
References
1. Deshpande BR, Katz JN, Solomon DH, et al. Number of persons
with symptomatic knee osteoarthritis in the US: impact of race
and ethnicity, age, sex, and obesity. Arthritis Care Res (Hoboken)
2016; 68:1743–1750.
2. Bannuru RR, Schmid CH, Kent DM, Vaysbrot EE, Wong JB,
McAlindon TE. Comparative effectiveness of pharmacologic inter-
ventions for knee osteoarthritis: a systematic review and network
meta-analysis. Ann Intern Med 2015; 162:46–54.
3. Gregori D, Giacovelli G, Minto C, et al. Association of pharmaco-
logical treatments with long-term pain control in patients with
knee osteoarthritis: a systematic review and meta-analysis. JAMA
2018; 320:2564–2579.
4. Rutjes AW, Jüni P, da Costa BR, Trelle S, Nüesch E,
Reichenbach S. Viscosupplementation for osteoarthritis of the
knee: a systematic review and meta-analysis. Ann Intern Med 2012;
157:180–191.
5. Bellamy N, Campbell J, Robinson V, Gee T, Bourne R, Wells G.
Viscosupplementation for the treatment of osteoarthritis of the
knee. Cochrane Database Syst Rev 2006; 2:CD005321.
6. Johansen M, Bahrt H, Altman RD, et al. Exploring reasons for the
observed inconsistent trial reports on intra-articular injections with
hyaluronic acid in the treatment of osteoarthritis: meta-regression
analyses of randomized trials. Semin Arthritis Rheum 2016; 46:34–48.
7. Strand V, McIntyre LF, Beach WR, Miller LE, Block JE. Safety
and efficacy of US-approved viscosupplements for knee osteoar-
thritis: a systematic review and meta-analysis of randomized,
saline-controlled trials. J Pain Res 2015; 8:217–228.
8. Maricar N, Parkes MJ, Callaghan MJ, Felson DT, O’Neill TW.
Where and how to inject the knee: a systematic review. Semin
Arthritis Rheum 2013; 43:195–203.
9. Daniels EW, Cole D, Jacobs B, Phillips SF. Existing evidence on
ultrasound-guided injections in sports medicine. Orthop J Sports
Med 2018; 6:2325967118756576.
10. Altman R, Fredericson M, Bhattacharyya SK, et al. Association
between hyaluronic acid injections and time-to-total knee replace-
ment surgery. J Knee Surg 2016; 29:564–570.
11. Altman R, Lim S, Steen RG, Dasa V. Hyaluronic acid injections
are associated with delay of total knee replacement surgery in
patients with knee osteoarthritis: evidence from a large US health
claims database. PLoS One 2015; 10:e0145776.
12. Delbarre A, Amor B, Bardoulat I, Tetafort A, Pelletier-Fleury N.
Do intra-articular hyaluronic acid injections delay total knee

replacement in patients with osteoarthritis: a Cox model analysis.
PLoS One 2017; 12:e0187227.
13. Ong KL, Anderson AF, Niazi F, Fierlinger AL, Kurtz SM,
Altman RD. Hyaluronic acid injections in Medicare knee osteoar-
thritis patients are associated with longer time to knee arthroplasty.
J Arthroplasty 2016; 31:1667–1673.
14. Waddell DD, Bricker DC. Total knee replacement delayed with
Hylan G-F 20 use in patients with grade IV osteoarthritis. J Manag
Care Pharm 2007; 13:113–121.
15. Shewale AR, Barnes CL, Fischbach LA, Ounpraseuth ST,
Painter JT, Martin BC. Comparison of low-, moderate-, and
J Ultrasound Med 2019
Lundstrom et al—Ultrasound-Guided Knee Viscosupplementation
high-molecular-weight hyaluronic acid injections in delaying time
to knee surgery. J Arthroplasty 2017; 32:2952-2957.e21.
16. Altman RD, Dasa V, Takeuchi J. Review of the mechanism of action
for Supartz FX in knee osteoarthritis. Cartilage 2018; 9:11–20.
17. Altman RD, Manjoo A, Fierlinger A, Niazi F, Nicholls M. The
mechanism of action for hyaluronic acid treatment in the osteoar-
thritic knee: a systematic review. BMC Musculoskelet Disord 2015;
16:321.
18. Bowman S, Awad ME, Hamrick MW, Hunter M, Fulzele S.
Recent advances in hyaluronic acid based therapy for osteoarthritis.
Clin Transl Med 2018; 7:6.
5