Introduction
A novel coronavirus disease (COVID-19) was identified in several patients living in
Wuhan, China, and first reported in December 2019 [35]. By early June 2020, the disease
had been diagnosed in 6.6 million people and 388,000 deaths had been reported in 215
countries. Mortality rates were high in some European countries; in Italy, the rapid
growth of COVID-19 cases led to shortages of hospital beds, ventilators, and personal
protective equipment (PPE) [2].
The USA has had the highest number of confirmed cases–nearly 6 million by late August
2020 [15]. The first US case was reported in January 2020 [12], and the first case
in New York State on March 1, 2020 [32]. As of May 24, 2020, New York State had the
highest number of confirmed cases in the nation (374,085) and deaths (24,079) [15].
The state had more confirmed cases than many countries, and within New York State,
New York City and its nearby counties of Westchester, Suffolk, and Nassau had 80%
of all confirmed COVID-19 cases in the state [15].
The experience in other countries, coupled with predictions that COVID-19 could overwhelm
local healthcare resources, led to a halt in all elective surgery. On March 7, 2020,
the Governor of New York State, Andrew Cuomo, declared a state of emergency after
89 cases were confirmed in the state and after predictions indicated that the hospital
capacity for beds, ICU beds, and ventilators would be insufficient to cope with the
demand of people affected by COVID-19. Hospital for Special Surgery (HSS) suspended
all elective outpatient visits and surgeries in mid-March 2020, a decision that had
last been taken after the terrorist attack on September 11, 2001. Our hospital’s decision
was soon followed by an executive order by the governor stopping all non-essential
medical and surgical care in hospitals, clinics, surgical centers, and outpatient
offices statewide.
During March, April, and May, HSS, which had been primarily devoted to elective treatment
of orthopedic and rheumatologic conditions, underwent a substantial transformation
and admitted COVID-19 patients and others affected by non-orthopedic conditions and
emergent orthopedic conditions. Non-critical COVID-19 patients were attended to in
a devoted medical unit, whereas critical COVID-19 patients received care in operating
rooms that had been transformed to an intensive care unit (ICU).
Included from the start was planning for the return to normal operations after the
emergency was resolved. This review narrates the steps taken by the Adult Reconstruction
and Joint Replacement (ARJR) service to frame and organize the return to non-emergent
orthopedic surgery. Our hope is that others can benefit from our experience while
dealing with this pandemic or other situations that require similar, drastic measures.
Brainstorming the Return to Non-Emergent Surgery
After cancelling all elective surgery, the ARJR service was divided into three groups
of attending surgeons and trainees. The groups functioned as autonomous cells to take
care of orthopedic emergencies in our own patients and those from nearby and affiliated
institutions that were fully devoted to the care of COVID-19 patients. Each group
would see emergencies and operate on a rotating schedule [6, 27].
Two weeks after halting elective care, a group of four attending arthroplasty surgeons
(FB, SH, MF, AGDV) were tasked with brainstorming ideas about returning to non-emergent
arthroplasty care, while substantial efforts were being made to expedite a telehealth
program that would allow patients to have access to limited, non-emergent medical
advice.
Meeting virtually twice a week, we worked on a model that would focus on re-initiating
non-urgent outpatient and inpatient care related to arthroplasty focusing our efforts
on seven areas that would cover most of our patient interactions: (1) the timing to
re-start non-emergent arthroplasty surgery; (2) COVID-19 screening and testing; (3)
the office visit and ancillary testing; (4) pre-surgical screening (PSS); (5) the
operating room; (6) the hospital stay; and (7) recovery after discharge. We conceived
a plan that would minimize the risk of both health care workers (HCW) and patients
of contracting COVID-19.
Our plan relied on the need to adapt to a very dynamic situation. In a short period
of time, the medical community gained knowledge on this new pathogen and its resulting
disease. Our group organized weekly conference calls with colleagues in China, Singapore,
and South Korea, areas that had experienced an earlier increase in the number of cases
and had put into practice different containment strategies [19, 20]. Their experience,
and that of colleagues in European countries that also experienced a high number of
cases like Italy and Spain, proved to be very useful.
The Timing to Return to Non-Emergent Care
The daily monitoring of the number of new COVID-19 cases, the number of new hospital
admissions, the number of admissions to the ICU, the number of hospital discharges,
and deaths associated with COVID-19 in New York State proved to be crucial to timing
the re-start of non-emergent surgery. Guided by the governor, New York State institutions
were directed to develop step-by-step protocols for re-opening, including the following
guidelines for non-emergent surgery:
New infections: In accordance with Centers for Disease Control and Prevention guidelines,
health care facilities considering re-opening should have had at least 14 days of
decline in total net hospitalizations and deaths on a 3-day rolling average. Regions
with few COVID-19 cases could not exceed 15 net new total hospitalizations or five
new deaths on a 3-day rolling average. In addition, the region must have had fewer
than two new COVID-19 patients admitted per 100,000 residents per day.
Health care capacity: Every region must have had the health care capacity to handle
a potential resurgence in cases. Hospitals in the region were required to reserve
at least 30% of their hospital and ICU beds and to stockpile at least 90 days of personal
protective equipment (PPE).
Testing: Facilities were ordered to implement testing regimens that prioritized symptomatic
individuals and those who had come into contact with them and to conduct frequent
tests of frontline and essential workers. Each region had to have the capacity to
conduct 30 diagnostic tests for every 1000 residents per month.
Contact tracing: Regions had to ensure a baseline of 30 contact tracers for every
100,000 residents, with additional tracers based on the projected number of cases
in the region.
Isolation: Facilities had to present plans to have rooms available for patients testing
positive for COVID-19 and who could self-isolate as part of the re-opening process.
These requirements were overseen by the state government. The number of cases nationwide
and particularly in New York State, coupled with the lack of pharmacologic treatment
or prevention, made officials, hospitals, and the public realize that life would not
be the same until an effective treatment or vaccine were available. The expression
“return to new normal” was coined.
While the state government guided the timing to return to elective surgeries, by May
2020, hospitals started to ease criteria for emergency medical visits and surgeries.
At HSS beginning in early May 2020, guidelines from the American College of Surgeons
[30] and the Centers for Medicare and Medicaid Services [7] for medical care including
surgery were offered first to emergent, followed by urgent, and then priority cases
(Table 1).
Table 1
Classification of essential and elective surgeries
Category
General description
Arthroplasty surgery
Essential
Emergent
Need for immediate care. Delay can be life- of limb-threatening or result in long
term functional disability.
THA for hip fracture, peri-prosthetic fracture, prosthetic joint infection, catastrophic
mechanical failure of implant, manipulation under anesthesia, evacuation of hematoma.
Urgent
Condition that if left untreated can result in sub-optimal outcome, aggravation of
underlying medical condition, or chronic opioid dependence.
Pending peri-prosthetic fracture, some prosthetic joint infection, recurrent dislocation,
advanced osteonecrosis with collapse and bone loss, rapidly progressive OA, some reimplantation.
Priority
Condition that if left untreated will result in sub-optimal outcome for a variety
of medical or socioeconomic factors
Some primary and revision arthroplasty, some revisions and reimplantation.
Elective
Low risk
Procedure that will improve patient’s function. There is no time constraint on its
optimal outcome. Low-risk patients only.
Elective primary and revision arthroplasty. Single stage bilateral arthroplasty.
All patients
Procedure that will improve patient’s function. There is no time constraint on its
optimal outcome. Low-risk patients only.
Elective arthroplasty.
Because the outcome of elective surgery in patients who develop COVID-19 is unclear,
strict safety guidelines were established not authorizing urgent and priority procedures
in patients at increased risk for COVID-19 complications. In a case series describing
four patients in the first weeks of the COVID-19 outbreak who underwent elective cholecystectomy,
hernia repair, gastric bypass, and hysterectomy, peri-operative complications were
significant [1]. Three of the four patients died from complications typical of rapid
progression of COVID-19. Similarly, Li et al., in reporting a death rate of 20.6%
in 34 patients operated on electively during the COVID-19 outbreak, suggested that
surgery may exacerbate and accelerate disease progression [18].
HSS established safety criteria for the careful selection of patients who would be
eligible for urgent and priority surgery. These criteria were modified and eased during
the “return to new normal” (Table 2). The ultimate goal was to select patients who
were at low risk for procedure-related complications, would require short hospitalizations,
and had a low risk of death should they contract COVID-19 post-operatively [10].
Table 2
Risk stratification for patients undergoing surgery
Number
1
COVID-19 negative
2
Age < 70 years (later expanded to age < 60 at the end of May 2020)
3
ASA class 1 or 2
4
Body mass index < 40
5
Non-smoker
6
Low risk for complex pain management challenges
7
Opioid use < 6 months
8
No history of substance use disorder, including active/current licit or illicit substance
abuse
9
No intrathecal pump
10
No history of or current buprenorphine use
11
No history of coronary artery disease, heart failure, valvular heart disease, pulmonary
disease, immunosuppressive disease or therapy
12
Diabetes: none or HbA1c < 7 (later expanded to < 8 at the end of May 2020)
13
Chronic kidney disease class 1–2
14
Expected hospitalization ≤ 48h
15
Not expected to need inpatient pain consult
16
Not expected to need blood transfusion
17
Independent functional status pre-operatively
18
Expected discharge to home (e.g., RAPT ≥ 9)
19
Ability to participate in remote/telerehabilitation
RAPT Risk Assessment and Prediction Tool, ASA American Society of Anesthesiologists
Screening and Testing for COVID-19
Clinical screening was implemented for all patients coming to our hospital, which
included measuring body temperature and asking a series of questions assessing for
fever, dry cough, sore throat, anosmia, tiredness, continuous rash, and gastrointestinal
symptoms [16] experienced in the prior 2 weeks. Other institutions implemented a full
physical evaluation for all patients [28, 29]. By the time hydroxychloroquine was
being evaluated as a preventive and potentially therapeutic agent, some institutions
inquired about the history of receiving prophylactic medication against COVID-19 [5].
One cornerstone in the effort to control the disease and return to safe elective surgery
is accurate, rapid testing. At the beginning of the pandemic, testing was recommended
only for patients with symptoms compatible with COVID-19. However, as the hospital
began the return to new normal, all patients requiring admission were tested. All
patients undergoing emergent, urgent, priority, or elective surgery underwent testing
within 24 h of admission. This included those without symptoms of COVID-19, as asymptomatic
contacts may transmit the disease during the incubation period [26]. Only COVID-19-negative
patients could undergo urgent, priority, and elective surgery [16].
The most common recommendation for testing is a reverse transcriptase polymerase chain
reaction (RT-PCR) test via nasopharyngeal swab, which is considered highly specific
for SARS-CoV-2 [8, 36], but several studies had reported a significant false-negative
rate [31, 34] with this test. Our hospital procured the equipment and reagents to
perform in-house testing in hopes of reducing the false-negative incidence.
In addition to the PCR test, our protocol included testing for SARS-CoV-2 antibodies
that would have indicated prior exposure or resolved infection. This information is
essential to identify patients who had COVID-19 and who might be at an increased risk
for complications post-operatively. A new serology test by Roche administered 14 days
after COVID-19 PCR confirmation reported a specificity of 99.8% and a sensitivity
of 100% [25]. Our early experience with overflow transfers to our hospital from other
institutions for treatment of orthopedic injuries suggested that some asymptomatic
patients without a history of COVID-19 presented with hypoxemia or even infiltrates
on screening chest radiographs. As a result, the pre-operative screening process for
elective surgery during this period mandated a screening chest X-ray and assessment
of oxygenation using pulse oximetry. Pulmonary computed tomography (CT) had been recommended
by some authors for screening [13, 14, 33] due to the possibility of a false-negative
RT-PCR test; however, we relied on pulse oximetry and chest radiographs.
The Office Environment for Urgent and Non-Urgent Office Visits
Immediately after the shutdown in mid-March 2020, telehealth was implemented. Online
training was made available, and over 210 physicians were trained. The telemedicine
platform went live in 4 weeks. An additional 135 rehabilitation specialists (physical,
occupational, speech, and performance therapists) were trained in 3 weeks. Prior to
the pandemic, physical therapists held on average 20 telehealth visits per week as
part of post-operative monitoring. During the pandemic, we reached over 1000 visits
per day between physicians and therapists. Telerehabilitation with patients who had
undergone arthroplasty in the weeks prior to the shutdown proved very useful. In our
area, total knee replacement patients rely heavily on in-person physical therapy services
after discharge. Telemedicine allowed us to conduct routine post-operative checks
that we would otherwise have conducted in person.
From mid-March to early June 2020, outpatient clinics to assess and treat patients
with emergencies related to arthroplasty continued to function. These clinics were
available in the main campus in Manhattan and in satellite locations in White Plains,
Uniondale, and Stamford, areas less affected by COVID-19 and where COVID-19-negative
patients could feel more comfortable receiving essential care.
In addition to the outpatient emergency clinic, beginning in mid-May 2020, a limited
office schedule was implemented at the main campus and the satellites to provide care
to patients with urgent and priority conditions (Table 2). Surgeons saw no more than
three patients per hour (24 patients in an 8-h period). This limited the number of
staff coming in to work, maintained social distancing, and allowed for cleaning of
exam rooms between patients. Each surgeon saw patients once a week in a predetermined
location and was encouraged to use the satellite offices.
We implemented a strict clinical screening for all patients at our facilities. Face
masks and hand sanitizing was provided, and no visitors were allowed. An appointment
was required, no walk-ins were allowed, and questionnaires on COVID-19 symptoms were
completed for all appointments.
Our elective outpatient clinics remained closed until early June 2020, re-opening
on a similar limited schedule of days and number of daily appointments. During these
phases, our office staff worked remotely. In order to obtain valuable information
that would shorten the face-to-face time in the office, we strongly encouraged physicians
to perform telehealth visits on new patients prior to their in-person visit. We predict
that in the future, telehealth for routine follow-ups and patients coming from long
distances will be a routine part of our practices.
In addition, we facilitate the office visit by pre-scheduling any needed X-rays, magnetic
resonance imaging (MRI), and testing. Patients requiring imaging before their visit
were given an appointment and asked to wait until they received a text message to
enter the building and the radiology suite. Patients requiring an X-ray, MRI, or special
procedure without an office visit were placed on a list and contacted by radiology
for a scheduled visit at times when the facilities were less busy.
Prioritizing Patients for Non-Urgent Surgery and the Selection Criteria
In prioritizing patients for non-emergency arthroplasty, and in view of the backlog
generated by the halt in non-urgent surgery, patients were assessed for the severity
of their symptoms, disability, and radiographic findings. Surgeons were asked to create
priority lists for their practices using a five-level scale (Table 1). Such priority
rating systems are routinely used in national health care systems [3, 9]. In May 2020,
essential surgeries could proceed, providing the hospital had adequate resources and
protocols in place for them to be performed safely.
The second set of criteria was based on individual risk factors (Table 2). COVID-19
posed substantial additional risk to older patients with comorbidities such as coronary
artery disease, chronic obstructive pulmonary disease, or uncontrolled diabetes [11].
Medical safety guidelines were developed, and patients were risk stratified [21, 22].
The totality of the patient’s medical condition and need for surgery were individualized.
Guidelines were helpful for decision-making. Additional considerations included social
and rehabilitation factors. Patients who required inpatient rehabilitation and prolonged
hospitalization were deemed inappropriate for the early phase of non-emergent surgery.
It was important to document the appropriateness for surgery and to obtain informed
consent for procedures during this period. We established a surgical oversight committee
to review case selection and documentation for our surgical cases.
The Patient’s Journey While Preparing for Arthroplasty
Our goal was to streamline the process of PSS and offer patients the possibility of
completing tasks online rather than in the hospital. The idea was to shorten the time
required for PSS. The one-and-a-half-hour, in-person, pre-operative education class
for patients undergoing THA and TKA was converted to online education modules patients
took before their PSS visit. Likewise, in a pre-operative telerehabilitation visit,
the most important pre- and post-operative exercises and precautions were discussed,
while preparing patients for a post-operative therapy plan that would include telerehabilitation
and/or in-person therapy. In-person discharge planning with a social worker was replaced
by a pre-operative discharge planning call, during which a case manager would confirm
the plan, eliminating unnecessary delays in discharge after surgery.
PSS was performed either at the main institution or at the satellites. Satellites
had previously been utilized for pre-operative testing rarely. In order to leverage
the benefits of the satellite offices, hospitalists assigned clinic days at the satellites
facilitated the pre-operative evaluations. All patients had antibody (IgG) testing
for SARS-CoV-2 within a week of surgery. If the antibody test was positive, surgery
was scheduled for the last case of the day.
Anesthesiologists were encouraged to call patients the day before surgery, to shorten
the face-to-face dialog prior to surgery. This ensured that questions could be answered
prior to coming into the hospital and maximized professional distancing.
In order to further minimize waiting times, pre-operative radiographs, MRIs, or CT
scans were scheduled along with other exams required for PSS. Patients who had only
a telehealth visit before scheduling surgery would also see their surgeon on PSS day
for a complete physical exam, evaluation of radiographs, and discussion of questions
not covered via telehealth. Designed to minimize face-to-face contact, many of these
changes have streamlined the patient experience and are expected to continue in the
post-pandemic period.
The Surgeries and the Operative Environment
As our region began to move beyond the crisis, indications for non-urgent arthroplasty
were expanded. Our hospital took a thoughtful and staged approach, while complying
with local and federal guidelines. In early May 2020, urgent and priority arthroplasties
in patients who had been waiting almost 2 months were progressively allowed. These
included a limited number of patients with advanced joint disease and concomitant
bone loss, those with grossly loose implants and pending peri-prosthetic fractures,
those who had undergone removal of an infected joint replacement and required reimplantation,
and those with poorly controlled pain (Table 1).
As the hospital started to expand surgical indications, the surgical schedule remained
substantially reduced. This allowed safe distancing for the protection of patients
and HCWs and gave the team time to comply with in-hospital security measures and testing.
All patients and staff were screened daily upon entry and were required to wear PPE.
Social distancing was maintained between patients; in order to do so, schedules were
modified to accommodate the additional space and staff requirements. All patients
underwent PCR swab testing upon admission and prior to surgery [2, 11, 23]. The turnaround
time for COVID-19 PCR testing was 60 min and was incorporated into the workflow. Facilities
that cannot do rapid testing should try to have testing performed as close to surgery
as possible.
New operating room (OR) safety protocols were developed for universal use of PPE and
proper waiting times. Rigorous adherence to these newer safety protocols should take
priority over efficiency in the early expansion of surgical indications.
Careful consideration was taken to create the appropriate operating room airflow and
sterile environment. The International Consensus Group (ICM) recommended a normal
positive-pressure room modality to decrease efflux such as in-room air filters or
negative-pressure antechambers [24]. The European Knee Society [11] and ICM [24] also
recommend air flow with a minimum of 20 air changes per hour. At our institution,
air circulation is via laminar flow with high-efficiency particulate air (HEPA) changed
20 times per hour through the ceiling, coming down on the patient from above and then
being drawn out of the room through two exhaust vents on opposite sides of the room.
The majority of arthroplasty surgery at our hospital was and is performed with neuraxial
anesthesia. For potentially aerosolizing procedures (intubation or extubation, for
example), indispensable OR personnel remained in the room utilizing PPE that include
N95 masks and goggles, and the OR doors were kept closed for 20 min to allow for complete
air turnover, thereby protecting the hallways. In addition, portable HEPA filters
have been installed at the head of each operating table to immediately filter air
coming from the patient.
The Post-Operative Hospital Stay
Patients undergoing arthroplasty who required a short hospitalization or same-day
discharge were preferred in the early phases. This was carefully discussed with the
patients, and the conditions for an early, safe discharge were optimized. Ambulatory
arthroplasty has been a growing trend in the USA for several years. Advocates have
emphasized the safety of outpatient procedures and the avoidance of inpatients risks
[4]. In the era of COVID-19, same-day or short-stay surgery is especially appealing.
In spite of hospitals’ extensive efforts to minimize COVID-19 risks to patients, other
patients and/or HCW potentially transmit the virus in spite of testing negative for
COVID-19 [17]. Patients at home are likely to have fewer personal interactions, limiting
the exposure risk. During this period all hospital rooms were single occupancy. During
early phases of return to normal, no visitors or visiting hours were allowed for adult
patients.
In conclusion, the COVID-19 crisis had a major impact on elective surgeries including
arthroplasty. As our hospital emerged from the crisis, we made adaptative responses
in care that we implemented on the basis of understanding the spread of COVID-19 in
our area; our patients’ characteristics, social and living conditions; and their priorities
in undergoing different kinds of surgery. We leveraged pre-crisis structures and procedures
and created others to allow the progressive return to non-urgent arthroplasty.
Electronic Supplementary Material
ESM 1
(PDF 1224 kb)
ESM 2
(PDF 1225 kb)
ESM 3
(PDF 1225 kb)
ESM 4
(PDF 1225 kb)
ESM 5
(PDF 1224 kb)
ESM 6
(PDF 1225 kb)
ESM 7
(PDF 1225 kb)